1 .. This work is licensed under a Creative Commons Attribution 4.0 International
3 .. http://creativecommons.org/licenses/by/4.0
4 .. (c) OPNFV, 2017 Intel Corporation.
6 ***********************************************
7 Yardstick Test Case Description: NSB PROX L2FWD
8 ***********************************************
10 +-----------------------------------------------------------------------------+
11 |NSB PROX test for NFVI characterization |
13 +--------------+--------------------------------------------------------------+
14 |test case id | tc_prox_{context}_l2fwd-{port_num} |
16 | | * context = baremetal or heat_context; |
17 | | * port_num = 2 or 4; |
19 +--------------+--------------------------------------------------------------+
20 |metric | * Network Throughput; |
21 | | * TG Packets Out; |
22 | | * TG Packets In; |
23 | | * VNF Packets Out; |
24 | | * VNF Packets In; |
25 | | * Dropped packets; |
27 +--------------+--------------------------------------------------------------+
28 |test purpose | The PROX L2FWD test has 3 types of test cases: |
29 | | L2FWD: The application will take packets in from one port |
30 | | and forward them unmodified to another port |
31 | | L2FWD_Packet_Touch: The application will take packets in |
32 | | from one port, update src and dst MACs and forward them to |
34 | | L2FWD_Multi_Flow: The application will take packets in |
35 | | from one port, update src and dst MACs and forward them to |
36 | | another port. This test case exercises the softswitch |
37 | | with 200k flows. |
39 | | The above test cases are implemented for baremetal and heat |
40 | | context for 2 port and 4 port configuration. |
42 +--------------+--------------------------------------------------------------+
43 |configuration | The L2FWD test cases are listed below: |
45 | | * tc_prox_baremetal_l2fwd-2.yaml |
46 | | * tc_prox_baremetal_l2fwd-4.yaml |
47 | | * tc_prox_baremetal_l2fwd_pktTouch-2.yaml |
48 | | * tc_prox_baremetal_l2fwd_pktTouch-4.yaml |
49 | | * tc_prox_baremetal_l2fwd_multiflow-2.yaml |
50 | | * tc_prox_baremetal_l2fwd_multiflow-4.yaml |
51 | | * tc_prox_heat_context_l2fwd-2.yaml |
52 | | * tc_prox_heat_context_l2fwd-4.yaml |
53 | | * tc_prox_heat_context_l2fwd_pktTouch-2.yaml |
54 | | * tc_prox_heat_context_l2fwd_pktTouch-4.yaml |
55 | | * tc_prox_heat_context_l2fwd_multiflow-2.yaml |
56 | | * tc_prox_heat_context_l2fwd_multiflow-4.yaml |
58 | | Test duration is set as 300sec for each test. |
59 | | Packet size set as 64 bytes in traffic profile |
60 | | These can be configured |
62 +--------------+--------------------------------------------------------------+
64 | | PROX is a DPDK application that can simulate VNF workloads |
65 | | and can generate traffic and used for NFVI characterization |
67 +--------------+--------------------------------------------------------------+
68 |applicability | The PROX L2FWD test cases can be configured with different: |
71 | | * test durations; |
72 | | * tolerated loss; |
74 | | Default values exist. |
76 +--------------+--------------------------------------------------------------+
77 |pre-test | For Openstack test case image (yardstick-samplevnfs) needs |
78 |conditions | to be installed into Glance with Prox and Dpdk included in |
81 | | For Baremetal tests cases Prox and Dpdk must be installed in |
82 | | the hosts where the test is executed. The pod.yaml file must |
83 | | have the necessary system and NIC information |
85 +--------------+--------------------------------------------------------------+
86 |test sequence | description and expected result |
88 +--------------+--------------------------------------------------------------+
89 |step 1 | For Baremetal test: The TG and VNF are started on the hosts |
90 | | based on the pod file. |
92 | | For Heat test: Two host VMs are booted, as Traffic generator |
93 | | and VNF(L2FWD workload) based on the test flavor. |
95 +--------------+--------------------------------------------------------------+
96 |step 2 | Yardstick is connected with the TG and VNF by using ssh. |
97 | | The test will resolve the topology and instantiate the VNF |
98 | | and TG and collect the KPI's/metrics. |
100 +--------------+--------------------------------------------------------------+
101 |step 3 | The TG will send packets to the VNF. If the number of |
102 | | dropped packets is more than the tolerated loss the line |
103 | | rate or throughput is halved. This is done until the dropped |
104 | | packets are within an acceptable tolerated loss. |
106 | | The KPI is the number of packets per second for 64 bytes |
107 | | packet size with an accepted minimal packet loss for the |
108 | | default configuration. |
110 +--------------+--------------------------------------------------------------+
111 |step 4 | In Baremetal test: The test quits the application and unbind |
112 | | the dpdk ports. |
114 | | In Heat test: Two host VMs are deleted on test completion. |
116 +--------------+--------------------------------------------------------------+
117 |test verdict | The test case will achieve a Throughput with an accepted |
118 | | minimal tolerated packet loss. |
119 +--------------+--------------------------------------------------------------+