aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--docs/tag_documentation.rst12
-rw-r--r--resources/libraries/python/TrafficGenerator.py45
-rw-r--r--resources/libraries/robot/performance.robot42
-rwxr-xr-xresources/tools/t-rex/t-rex-stateless.py72
-rw-r--r--tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot492
-rw-r--r--tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot492
-rw-r--r--tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot492
-rw-r--r--tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot213
-rw-r--r--tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot213
-rw-r--r--tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot213
10 files changed, 2260 insertions, 26 deletions
diff --git a/docs/tag_documentation.rst b/docs/tag_documentation.rst
index 1cb8147c7e..de97c6c9e9 100644
--- a/docs/tag_documentation.rst
+++ b/docs/tag_documentation.rst
@@ -97,3 +97,15 @@ PDR
NDR
Non Drop Rate evaluation of results. Loss acceptance of dropped packets is
set to zero lost packets.
+
+Scaling TAGs
+------------
+
+FIB_20K
+ 2x10,000 entries in single fib table
+
+FIB_200K
+ 2x100,000 entries in single fib table
+
+FIB_2M
+ 2x1,000,000 entries in single fib table
diff --git a/resources/libraries/python/TrafficGenerator.py b/resources/libraries/python/TrafficGenerator.py
index 6f642c72d0..a640c1d80a 100644
--- a/resources/libraries/python/TrafficGenerator.py
+++ b/resources/libraries/python/TrafficGenerator.py
@@ -346,6 +346,51 @@ class TrafficGenerator(object):
_p0, _p1, _async, _latency,
warmup_time),
timeout=int(duration)+60)
+ elif traffic_type in ["3-node-IPv4-dst-10000"]:
+ (ret, stdout, stderr) = ssh.exec_command(
+ "sh -c '{0}/resources/tools/t-rex/t-rex-stateless.py "
+ "--duration={1} -r {2} -s {3} "
+ "--p{4}_src_start_ip 10.0.0.1 "
+ "--p{4}_dst_start_ip 20.0.0.0 "
+ "--p{4}_dst_end_ip 20.0.39.15 "
+ "--p{5}_src_start_ip 20.0.0.1 "
+ "--p{5}_dst_start_ip 10.0.0.0 "
+ "--p{5}_dst_end_ip 10.0.39.15 "
+ "{6} {7} --warmup_time={8}'".format(Constants.REMOTE_FW_DIR,
+ duration, rate, framesize,
+ _p0, _p1, _async, _latency,
+ warmup_time),
+ timeout=int(duration)+60)
+ elif traffic_type in ["3-node-IPv4-dst-100000"]:
+ (ret, stdout, stderr) = ssh.exec_command(
+ "sh -c '{0}/resources/tools/t-rex/t-rex-stateless.py "
+ "--duration={1} -r {2} -s {3} "
+ "--p{4}_src_start_ip 10.0.0.1 "
+ "--p{4}_dst_start_ip 20.0.0.0 "
+ "--p{4}_dst_end_ip 20.1.134.159 "
+ "--p{5}_src_start_ip 20.0.0.1 "
+ "--p{5}_dst_start_ip 10.0.0.0 "
+ "--p{5}_dst_end_ip 10.1.134.159 "
+ "{6} {7} --warmup_time={8}'".format(Constants.REMOTE_FW_DIR,
+ duration, rate, framesize,
+ _p0, _p1, _async, _latency,
+ warmup_time),
+ timeout=int(duration)+60)
+ elif traffic_type in ["3-node-IPv4-dst-1000000"]:
+ (ret, stdout, stderr) = ssh.exec_command(
+ "sh -c '{0}/resources/tools/t-rex/t-rex-stateless.py "
+ "--duration={1} -r {2} -s {3} "
+ "--p{4}_src_start_ip 10.0.0.1 "
+ "--p{4}_dst_start_ip 20.0.0.0 "
+ "--p{4}_dst_end_ip 20.15.66.63 "
+ "--p{5}_src_start_ip 20.0.0.1 "
+ "--p{5}_dst_start_ip 10.0.0.0 "
+ "--p{5}_dst_end_ip 10.15.66.63 "
+ "{6} {7} --warmup_time={8}'".format(Constants.REMOTE_FW_DIR,
+ duration, rate, framesize,
+ _p0, _p1, _async, _latency,
+ warmup_time),
+ timeout=int(duration)+60)
elif traffic_type in ["3-node-IPv6"]:
(ret, stdout, stderr) = ssh.exec_command(
"sh -c '{0}/resources/tools/t-rex/t-rex-stateless.py "
diff --git a/resources/libraries/robot/performance.robot b/resources/libraries/robot/performance.robot
index f9fbfc7b5b..78cdd8b845 100644
--- a/resources/libraries/robot/performance.robot
+++ b/resources/libraries/robot/performance.robot
@@ -319,6 +319,48 @@
| | dut2_v4.set_route | 10.10.10.0 | 24 | 1.1.1.1 | ${dut2_if1}
| | All Vpp Interfaces Ready Wait | ${nodes}
+| Scale IPv4 forwarding initialized in a 3-node circular topology
+| | [Documentation]
+| | ... | Custom setup of IPv4 topology with scalability of ip routes on all
+| | ... | DUT nodes in 3-node circular topology
+| | ...
+| | ... | *Arguments:*
+| | ... | - ${count} - IP route count. Type: integer
+| | ...
+| | ... | *Return:*
+| | ... | - No value returned
+| | ...
+| | ... | *Example:*
+| | ...
+| | ... | \| Scale IPv4 forwarding initialized in a 3-node circular topology \
+| | ... | \| 100000 \|
+| | [Arguments] | ${count}
+| | Set Interface State | ${dut1} | ${dut1_if1} | up
+| | Set Interface State | ${dut1} | ${dut1_if2} | up
+| | Set Interface State | ${dut2} | ${dut2_if1} | up
+| | Set Interface State | ${dut2} | ${dut2_if2} | up
+| | ${tg1_if1_mac}= | Get Interface MAC | ${tg} | ${tg_if1}
+| | ${tg1_if2_mac}= | Get Interface MAC | ${tg} | ${tg_if2}
+| | ${dut1_if2_mac}= | Get Interface MAC | ${dut1} | ${dut1_if2}
+| | ${dut2_if1_mac}= | Get Interface MAC | ${dut2} | ${dut2_if1}
+| | Add arp on dut | ${dut1} | ${dut1_if1} | 1.1.1.1 | ${tg1_if1_mac}
+| | Add arp on dut | ${dut1} | ${dut1_if2} | 2.2.2.2 | ${dut2_if1_mac}
+| | Add arp on dut | ${dut2} | ${dut2_if1} | 2.2.2.1 | ${dut1_if2_mac}
+| | Add arp on dut | ${dut2} | ${dut2_if2} | 3.3.3.1 | ${tg1_if2_mac}
+| | IP addresses are set on interfaces | ${dut1} | ${dut1_if1} | 1.1.1.2 | 30
+| | IP addresses are set on interfaces | ${dut1} | ${dut1_if2} | 2.2.2.1 | 30
+| | IP addresses are set on interfaces | ${dut2} | ${dut2_if1} | 2.2.2.2 | 30
+| | IP addresses are set on interfaces | ${dut2} | ${dut2_if2} | 3.3.3.2 | 30
+| | Vpp Route Add | ${dut1} | 10.0.0.0 | 32 | 1.1.1.1 | ${dut1_if1}
+| | ... | count=${count}
+| | Vpp Route Add | ${dut1} | 20.0.0.0 | 32 | 2.2.2.2 | ${dut1_if2}
+| | ... | count=${count}
+| | Vpp Route Add | ${dut2} | 10.0.0.0 | 32 | 2.2.2.1 | ${dut2_if1}
+| | ... | count=${count}
+| | Vpp Route Add | ${dut2} | 20.0.0.0 | 32 | 3.3.3.1 | ${dut2_if2}
+| | ... | count=${count}
+| | All Vpp Interfaces Ready Wait | ${nodes}
+
| IPv6 forwarding initialized in a 3-node circular topology
| | [Documentation]
| | ... | Set UP state on VPP interfaces in path on nodes in 3-node circular
diff --git a/resources/tools/t-rex/t-rex-stateless.py b/resources/tools/t-rex/t-rex-stateless.py
index 04f7a25294..9ee34f6c22 100755
--- a/resources/tools/t-rex/t-rex-stateless.py
+++ b/resources/tools/t-rex/t-rex-stateless.py
@@ -194,29 +194,45 @@ def create_streams(traffic_options, frame_size=64):
p2_src_end_ip = traffic_options['p2_src_end_ip']
p2_dst_start_ip = traffic_options['p2_dst_start_ip']
+ p1_dst_end_ip = traffic_options['p1_dst_end_ip']
+ p2_dst_end_ip = traffic_options['p2_dst_end_ip']
+
base_pkt_a = Ether()/IP(src=p1_src_start_ip, dst=p1_dst_start_ip, proto=61)
base_pkt_b = Ether()/IP(src=p2_src_start_ip, dst=p2_dst_start_ip, proto=61)
- # The following code applies raw instructions to packet (IP src increment).
- # It splits the generated traffic by "ip_src" variable to cores and fix
- # IPv4 header checksum.
- vm1 = STLScVmRaw([STLVmFlowVar(name="src",
- min_value=p1_src_start_ip,
- max_value=p1_src_end_ip,
- size=4, op="inc"),
- STLVmWrFlowVar(fv_name="src", pkt_offset="IP.src"),
- STLVmFixIpv4(offset="IP"),
- ], split_by_field="src")
- # The following code applies raw instructions to packet (IP src increment).
- # It splits the generated traffic by "ip_src" variable to cores and fix
- # IPv4 header checksum.
- vm2 = STLScVmRaw([STLVmFlowVar(name="src",
- min_value=p2_src_start_ip,
- max_value=p2_src_end_ip,
- size=4, op="inc"),
- STLVmWrFlowVar(fv_name="src", pkt_offset="IP.src"),
- STLVmFixIpv4(offset="IP"),
- ], split_by_field="src")
+ # The following code applies raw instructions to packet (IP src/dst
+ # increment). It splits the generated traffic by "ip_src"/"ip_dst" variable
+ # to cores and fix IPv4 header checksum.
+ if p1_dst_end_ip and p2_dst_end_ip:
+ vm1 = STLScVmRaw([STLVmFlowVar(name="dst",
+ min_value=p1_dst_start_ip,
+ max_value=p1_dst_end_ip,
+ size=4, op="inc"),
+ STLVmWrFlowVar(fv_name="dst", pkt_offset="IP.dst"),
+ STLVmFixIpv4(offset="IP"),
+ ], split_by_field="dst")
+ vm2 = STLScVmRaw([STLVmFlowVar(name="dst",
+ min_value=p2_dst_start_ip,
+ max_value=p2_dst_end_ip,
+ size=4, op="inc"),
+ STLVmWrFlowVar(fv_name="dst", pkt_offset="IP.dst"),
+ STLVmFixIpv4(offset="IP"),
+ ], split_by_field="dst")
+ else:
+ vm1 = STLScVmRaw([STLVmFlowVar(name="src",
+ min_value=p1_src_start_ip,
+ max_value=p1_src_end_ip,
+ size=4, op="inc"),
+ STLVmWrFlowVar(fv_name="src", pkt_offset="IP.src"),
+ STLVmFixIpv4(offset="IP"),
+ ], split_by_field="src")
+ vm2 = STLScVmRaw([STLVmFlowVar(name="src",
+ min_value=p2_src_start_ip,
+ max_value=p2_src_end_ip,
+ size=4, op="inc"),
+ STLVmWrFlowVar(fv_name="src", pkt_offset="IP.src"),
+ STLVmFixIpv4(offset="IP"),
+ ], split_by_field="src")
return create_streams_v46(base_pkt_a, base_pkt_b, vm1, vm2, frame_size)
@@ -451,24 +467,28 @@ def parse_args():
# help="Port 1 destination MAC address")
parser.add_argument("--p1_src_start_ip", required=True,
help="Port 1 source start IP address")
- parser.add_argument("--p1_src_end_ip", required=True,
+ parser.add_argument("--p1_src_end_ip",
+ default=False,
help="Port 1 source end IP address")
parser.add_argument("--p1_dst_start_ip", required=True,
help="Port 1 destination start IP address")
-# parser.add_argument("--p1_dst_end_ip",
-# help="Port 1 destination end IP address")
+ parser.add_argument("--p1_dst_end_ip",
+ default=False,
+ help="Port 1 destination end IP address")
# parser.add_argument("--p2_src_mac",
# help="Port 2 source MAC address")
# parser.add_argument("--p2_dst_mac",
# help="Port 2 destination MAC address")
parser.add_argument("--p2_src_start_ip", required=True,
help="Port 2 source start IP address")
- parser.add_argument("--p2_src_end_ip", required=True,
+ parser.add_argument("--p2_src_end_ip",
+ default=False,
help="Port 2 source end IP address")
parser.add_argument("--p2_dst_start_ip", required=True,
help="Port 2 destination start IP address")
-# parser.add_argument("--p2_dst_end_ip",
-# help="Port 2 destination end IP address")
+ parser.add_argument("--p2_dst_end_ip",
+ default=False,
+ help="Port 2 destination end IP address")
return parser.parse_args()
diff --git a/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..07634aa60f
--- /dev/null
+++ b/tests/perf/Long_IPv4_Fib_200k_Intel-X520-DA2.robot
@@ -0,0 +1,492 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG
+| ... | FIB_200K | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 100k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 100000
+
+*** Test Cases ***
+| TC01: 64B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC02: 64B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC07: 64B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC08: 64B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC13: 64B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC14: 64B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x100k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
diff --git a/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..543f1d9123
--- /dev/null
+++ b/tests/perf/Long_IPv4_Fib_20k_Intel-X520-DA2.robot
@@ -0,0 +1,492 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG
+| ... | FIB_20K | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 10k flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 10000
+
+*** Test Cases ***
+| TC01: 64B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC02: 64B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC07: 64B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC08: 64B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC13: 64B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC14: 64B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x10k - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
diff --git a/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot b/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..1944ba4684
--- /dev/null
+++ b/tests/perf/Long_IPv4_Fib_2M_Intel-X520-DA2.robot
@@ -0,0 +1,492 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_LONG
+| ... | FIB_2M | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *RFC2544: Pkt throughput IPv4 routing test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* TG finds and reports throughput NDR (Non Drop
+| ... | Rate) with zero packet loss tolerance or throughput PDR (Partial Drop
+| ... | Rate) with non-zero packet loss tolerance (LT) expressed in percentage
+| ... | of packets transmitted. NDR and PDR are discovered for different
+| ... | Ethernet L2 frame sizes using either binary search or linear search
+| ... | algorithms with configured starting rate and final step that determines
+| ... | throughput measurement resolution. Test packets are generated by TG on
+| ... | links to DUTs. TG traffic profile contains two L3 flow-groups
+| ... | (flow-group per direction, 1M flows per flow-group) with all packets
+| ... | containing Ethernet header, IPv4 header with IP protocol=61 and
+| ... | generated payload. MAC addresses are matching MAC addresses of the TG
+| ... | node interfaces. Incrementing of IP.dst (IPv4 destination address) field
+| ... | is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 1000000
+
+*** Test Cases ***
+| TC01: 64B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC02: 64B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC03: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC04: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC05: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC06: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | PDR
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC07: 64B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC08: 64B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC09: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC10: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC11: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC12: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 2threads 2cores 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC13: 64B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC14: 64B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 64 Byte frames
+| | ... | using binary search start at 10GE linerate, step 100kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR
+| | ${framesize}= | Set Variable | 64
+| | ${min_rate}= | Set Variable | 100000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_64B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC15: 1518B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC16: 1518B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 1518 Byte frames
+| | ... | using binary search start at 10GE linerate, step 10kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 1518
+| | ${min_rate}= | Set Variable | 10000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_1518B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
+
+| TC17: 9000B NDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find NDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find NDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+
+| TC18: 9000B PDR binary search - DUT IPv4 Fib 2x1M - 4threads 4cores 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Find PDR for 9000 Byte frames
+| | ... | using binary search start at 10GE linerate, step 5kpps, LT=0.5%.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | PDR | SKIP_PATCH
+| | ${framesize}= | Set Variable | 9000
+| | ${min_rate}= | Set Variable | 5000
+| | ${max_rate}= | Set Variable | ${10Ge_linerate_pps_9000B}
+| | ${binary_min}= | Set Variable | ${min_rate}
+| | ${binary_max}= | Set Variable | ${max_rate}
+| | ${threshold}= | Set Variable | ${min_rate}
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Find PDR using binary search and pps | ${framesize} | ${binary_min}
+| | ... | ${binary_max}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+| | ... | ${min_rate} | ${max_rate}
+| | ... | ${threshold}
+| | ... | ${glob_loss_acceptance}
+| | ... | ${glob_loss_acceptance_type}
diff --git a/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..845036bcfd
--- /dev/null
+++ b/tests/perf/Short_IPv4_Fib_200k_Intel-X520-DA2.robot
@@ -0,0 +1,213 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT
+| ... | FIB_200K | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x100k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 100k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 100000
+
+*** Test Cases ***
+| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 3.5mpps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 7.5mpps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 10.0mpps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x100k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
diff --git a/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..9ac8c81536
--- /dev/null
+++ b/tests/perf/Short_IPv4_Fib_20k_Intel-X520-DA2.robot
@@ -0,0 +1,213 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT
+| ... | FIB_20K | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x10k static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 10k flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 10000
+
+*** Test Cases ***
+| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 3.5mpps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 7.5mpps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 10.0mpps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x10k - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
diff --git a/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot b/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot
new file mode 100644
index 0000000000..7caa56a5b5
--- /dev/null
+++ b/tests/perf/Short_IPv4_Fib_2M_Intel-X520-DA2.robot
@@ -0,0 +1,213 @@
+# Copyright (c) 2016 Cisco and/or its affiliates.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+*** Settings ***
+| Resource | resources/libraries/robot/performance.robot
+| Force Tags | 3_NODE_SINGLE_LINK_TOPO | PERFTEST | HW_ENV | PERFTEST_SHORT
+| ... | FIB_2M | NIC_Intel-X520-DA2
+| Suite Setup | 3-node Performance Suite Setup with DUT's NIC model
+| ... | L3 | Intel-X520-DA2
+| Suite Teardown | 3-node Performance Suite Teardown
+| Test Setup | Setup all DUTs before test
+| Test Teardown | Run Keyword | Remove startup configuration of VPP from all DUTs
+| Documentation | *Reference NDR throughput IPv4 routing verify test cases*
+| ...
+| ... | *[Top] Network Topologies:* TG-DUT1-DUT2-TG 3-node circular topology
+| ... | with single links between nodes.
+| ... | *[Enc] Packet Encapsulations:* Eth-IPv4 for IPv4 routing.
+| ... | *[Cfg] DUT configuration:* DUT1 and DUT2 are configured with IPv4
+| ... | routing and 2x1M static IPv4 /32 route entries. DUT1 and DUT2 tested
+| ... | with 2p10GE NIC X520 Niantic by Intel.
+| ... | *[Ver] TG verification:* In short performance tests, TG verifies
+| ... | DUTs' throughput at ref-NDR (reference Non Drop Rate) with zero packet
+| ... | loss tolerance. Ref-NDR value is periodically updated acording to
+| ... | formula: ref-NDR = 0.9x NDR, where NDR is found in RFC2544 long
+| ... | performance tests for the same DUT confiiguration. Test packets are
+| ... | generated by TG on links to DUTs. TG traffic profile contains two L3
+| ... | flow-groups (flow-group per direction, 1M flows per flow-group) with
+| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61
+| ... | and generated payload. Incrementing of IP.dst (IPv4 destination address)
+| ... | field is applied to both streams.
+| ... | *[Ref] Applicable standard specifications:* RFC2544.
+
+*** Variables ***
+| ${rts_per_flow}= | 1000000
+
+*** Test Cases ***
+| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 3.5mpps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | Verify ref-NDR for 1518 Byte frames using single trial throughput
+| | ... | test. DUT runs IPv4 routing config with 1 thread, 1 phy core, 1
+| | ... | receive queue per NIC port.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 1thread 1core 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 1 thread, 1 phy core, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 1_THREAD_NOHTT_RXQUEUES_1 | SINGLE_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '1' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 7.5mpps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 2threads, 2cores, 1rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 2 threads, 2 phy cores, \
+| | ... | 1 receive queue per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 2_THREAD_NOHTT_RXQUEUES_1 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '2' worker threads and rxqueues '1' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 64 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 64
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 10.0mpps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 1518 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 1518
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 812743pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Add No Multi Seg to all DUTs
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}
+
+| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 Fib 2x1M - 4threads, 4cores, 2rxq
+| | [Documentation]
+| | ... | [Cfg] DUT runs IPv4 routing config with 4 threads, 4 phy cores, \
+| | ... | 2 receive queues per NIC port. [Ver] Verify ref-NDR for 9000 Byte
+| | ... | frames using single trial throughput test.
+| | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | NDR
+| | ${framesize}= | Set Variable | 9000
+| | ${duration}= | Set Variable | 10
+| | ${rate}= | Set Variable | 138580pps
+| | Given Add '4' worker threads and rxqueues '2' without HTT to all DUTs
+| | And Add PCI devices to DUTs from 3-node single link topology
+| | And Apply startup configuration on all VPP DUTs
+| | And Scale IPv4 forwarding initialized in a 3-node circular topology
+| | ... | ${rts_per_flow}
+| | Then Traffic should pass with no loss | ${duration} | ${rate}
+| | ... | ${framesize}
+| | ... | 3-node-IPv4-dst-${rts_per_flow}