aboutsummaryrefslogtreecommitdiffstats
path: root/tests/perf
diff options
context:
space:
mode:
authorpmikus <pmikus@cisco.com>2016-06-17 08:08:08 +0100
committerMiroslav Miklus <mmiklus@cisco.com>2016-07-28 11:02:14 +0000
commite3f6b21b75b5d579ce3080a1f173828fc59270bf (patch)
tree7c66e02c30243e0b4e7a4c9474836996628c5c12 /tests/perf
parent2eb74f5c5cb58fcb072af765ea67dd80ee6d8755 (diff)
CSIT-169 IPv4 Scale - performance
- Add libraries for ipv4 scale setup - Add test cases for testing 10k/100k/1M fib entries - Add t-rex profiles for large scale testing Change-Id: If085e0d834e5ffe5538ba76756e8b9d0e8255cb2 Signed-off-by: pmikus <pmikus@cisco.com>
Diffstat (limited to 'tests/perf')
-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
6 files changed, 2115 insertions, 0 deletions
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}