diff options
Diffstat (limited to 'tests/suites/performance')
| -rw-r--r-- | tests/suites/performance/Short_IPv4_Cop_Intel-X520-DA2.robot | 83 | ||||
| -rw-r--r-- | tests/suites/performance/Short_IPv4_Intel-X520-DA2.robot | 82 |
2 files changed, 109 insertions, 56 deletions
diff --git a/tests/suites/performance/Short_IPv4_Cop_Intel-X520-DA2.robot b/tests/suites/performance/Short_IPv4_Cop_Intel-X520-DA2.robot index 4066840300..ff7bdcf1c0 100644 --- a/tests/suites/performance/Short_IPv4_Cop_Intel-X520-DA2.robot +++ b/tests/suites/performance/Short_IPv4_Cop_Intel-X520-DA2.robot @@ -23,13 +23,32 @@ | 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 | Minimal throughput acceptance test cases +| Documentation | *Reference NDR throughput IPv4 whitelist 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, two static IPv4 /24 routes and IPv4 COP security whitelist +| ... | ingress /24 filter entries applied on links TG - DUT1 and DUT2 - TG. +| ... | 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, 253 flows per flow-group) with +| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61 +| ... | and generated payload. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Test Cases *** -| 1core VPP passes 64B frames through IPv4 forwarding with COP at 2x 3.5Mpps in 3-node topology +| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 whitelist - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 64B frames through IPv4 forwarding -| | ... | at 2x 3.5Mpps in 3-node topology +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 64 | | ${duration}= | Set Variable | 10 @@ -48,10 +67,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 1core VPP passes 1518B frames through IPv4 forwarding with COP at 2x 812,743pps in 3-node topology +| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 whitelist - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 1518B frames through IPv4 forwarding -| | ... | at 2x 812,743pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 1518 | | ${duration}= | Set Variable | 10 @@ -70,10 +90,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 1core VPP passes 9000B frames through IPv4 forwarding with COP at 2x 138,580pps in 3-node topology +| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 whitelist - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 9000B frames through IPv4 forwarding -| | ... | at 2x 138,580pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 9000 | | ${duration}= | Set Variable | 10 @@ -91,10 +112,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 64B frames through IPv4 forwarding with COP at 2x 6.6Mpps in 3-node topology +| TC04: Verify 64B ref-NDR at 2x 6.6Mpps - DUT IPv4 whitelist - 2thread 2core 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 64B frames through IPv4 forwarding -| | ... | at 2x 6.6Mpps in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 64 | | ${duration}= | Set Variable | 10 @@ -113,10 +135,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 1518B frames through IPv4 forwarding with COP at 2x 812,743pps in 3-node topology +| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 whitelist - 2thread 2core 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 1518B frames through IPv4 forwarding -| | ... | at 2x 812,743pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 1518 | | ${duration}= | Set Variable | 10 @@ -135,10 +158,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 9000B frames through IPv4 forwarding with COP at 2x 138,580pps in 3-node topology +| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 whitelist - 2thread 2core 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 9000B frames through IPv4 forwarding -| | ... | at 2x 138,580pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters 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 | | ${framesize}= | Set Variable | 9000 | | ${duration}= | Set Variable | 10 @@ -156,10 +180,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 64B frames through IPv4 forwarding with COP at 2x 10.0Mpps in 3-node topology +| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 whitelist - 4threads 4cores 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 64B frames through IPv4 -| | ... | forwarding at 2x 10.0Mpps in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \ +| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify +| | ... | ref-NDR for 64 Byte frames using single trial throughput test. | | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | | ${framesize}= | Set Variable | 64 | | ${duration}= | Set Variable | 10 @@ -178,10 +203,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 1518B frames through IPv4 forwarding with COP at 2x 812,743pps in 3-node topology +| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 whitelist - 4threads 4cores 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 1518B frames through IPv4 -| | ... | forwarding at 2x 812,743pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \ +| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify +| | ... | ref-NDR for 1518 Byte frames using single trial throughput test. | | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | | ${framesize}= | Set Variable | 1518 | | ${duration}= | Set Variable | 10 @@ -200,10 +226,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 9000B frames through IPv4 forwarding with COP at 2x 138,580pps in 3-node topology +| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 whitelist - 4threads 4cores 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 9000B frames through IPv4 -| | ... | forwarding at 2x 138,580pps (2x 10Gbps) in 3-node topology with COP. +| | ... | [Cfg] DUT runs IPv4 routing and whitelist filters config with \ +| | ... | 4 thread, 4 phy core, 2 receive queue per NIC port. [Ver] Verify +| | ... | ref-NDR for 9000 Byte frames using single trial throughput test. | | [Tags] | 4_THREAD_NOHTT_RXQUEUES_2 | MULTI_THREAD | | ${framesize}= | Set Variable | 9000 | | ${duration}= | Set Variable | 10 diff --git a/tests/suites/performance/Short_IPv4_Intel-X520-DA2.robot b/tests/suites/performance/Short_IPv4_Intel-X520-DA2.robot index bf8c3dd030..10a2cf4b21 100644 --- a/tests/suites/performance/Short_IPv4_Intel-X520-DA2.robot +++ b/tests/suites/performance/Short_IPv4_Intel-X520-DA2.robot @@ -25,13 +25,31 @@ | 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 | Minimal throughput acceptance test cases +| 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 two static IPv4 /24 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, 253 flows per flow-group) with +| ... | all packets containing Ethernet header, IPv4 header with IP protocol=61 +| ... | and generated payload. +| ... | *[Ref] Applicable standard specifications:* RFC2544. *** Test Cases *** -| 1core VPP passes 64B frames through IPv4 forwarding at 2x 3.5Mpps in 3-node topology +| TC01: Verify 64B ref-NDR at 2x 3.5Mpps - DUT IPv4 - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 64B frames through IPv4 forwarding -| | ... | at 2x 3.5Mpps in 3-node topology. +| | ... | [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 @@ -44,10 +62,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 1core VPP passes 1518B frames through IPv4 forwarding at 2x 812,743pps in 3-node topology +| TC02: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 1518B frames through IPv4 forwarding -| | ... | at 2x 812,743pps (2x 10Gbps) in 3-node topology. +| | ... | [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 @@ -60,10 +79,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 1core VPP passes 9000B frames through IPv4 forwarding at 2x 138,580pps in 3-node topology +| TC03: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 - 1thread 1core 1rxq | | [Documentation] -| | ... | VPP with 1 core should pass 9000B frames through IPv4 forwarding -| | ... | at 2x 138,580pps (2x 10Gbps) in 3-node topology. +| | ... | [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 @@ -75,10 +95,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 64B frames through IPv4 forwarding at 2x 7.5Mpps in 3-node topology +| TC04: Verify 64B ref-NDR at 2x 7.5Mpps - DUT IPv4 - 2threads, 2cores, 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 64B frames through IPv4 forwarding -| | ... | at 2x 7.5Mpps in 3-node topology. +| | ... | [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 @@ -91,10 +112,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 1518B frames through IPv4 forwarding at 2x 812,743pps in 3-node topology +| TC05: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 - 2threads, 2cores, 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 1518B frames through IPv4 forwarding -| | ... | at 2x 812,743pps (2x 10Gbps) in 3-node topology. +| | ... | [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 @@ -107,10 +129,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 2core VPP with rxqueues 1 passes 9000B frames through IPv4 forwarding at 2x 138,580pps in 3-node topology +| TC06: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 - 2threads, 2cores, 1rxq | | [Documentation] -| | ... | VPP with 2 cores should pass 9000B frames through IPv4 forwarding -| | ... | at 2x 138,580pps (2x 10Gbps) in 3-node topology. +| | ... | [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 @@ -122,10 +145,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 64B frames through IPv4 forwarding at 2x 10.0Mpps in 3-node topology +| TC07: Verify 64B ref-NDR at 2x 10.0Mpps - DUT IPv4 - 4threads, 4cores, 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 64B frames through IPv4 -| | ... | forwarding at 2x 10.0Mpps in 3-node topology. +| | ... | [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 @@ -138,10 +162,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 1518B frames through IPv4 forwarding at 2x 812,743pps in 3-node topology +| TC08: Verify 1518B ref-NDR at 2x 812.74kpps - DUT IPv4 - 4threads, 4cores, 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 1518B frames through IPv4 -| | ... | forwarding at 2x 812,743pps (2x 10Gbps) in 3-node topology. +| | ... | [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 @@ -154,10 +179,11 @@ | | Then Traffic should pass with no loss | ${duration} | ${rate} | | ... | ${framesize} | 3-node-IPv4 -| 4core VPP with rxqueues 2 passes 9000B frames through IPv4 forwarding at 2x 138,580pps in 3-node topology +| TC09: Verify 9000B ref-NDR at 2x 138.58kpps - DUT IPv4 - 4threads, 4cores, 2rxq | | [Documentation] -| | ... | VPP with 4 cores and rxqueues 2 should pass 9000B frames through IPv4 -| | ... | forwarding at 2x 138,580pps (2x 10Gbps) in 3-node topology. +| | ... | [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 |