1 .. This work is licensed under a Creative Commons Attribution 4.0 International
3 .. http://creativecommons.org/licenses/by/4.0
4 .. (c) OPNFV, Ericsson AB and others.
6 *************************************
7 Yardstick Test Case Description TC001
8 *************************************
10 .. _pktgen: https://www.kernel.org/doc/Documentation/networking/pktgen.txt
12 +-----------------------------------------------------------------------------+
13 |Network Performance |
15 +--------------+--------------------------------------------------------------+
16 |test case id | OPNFV_YARDSTICK_TC001_NETWORK PERFORMANCE |
18 +--------------+--------------------------------------------------------------+
19 |metric | Number of flows and throughput |
21 +--------------+--------------------------------------------------------------+
22 |test purpose | The purpose of TC001 is to evaluate the IaaS network |
23 | | performance with regards to flows and throughput, such as if |
24 | | and how different amounts of flows matter for the throughput |
25 | | between hosts on different compute blades. Typically e.g. |
26 | | the performance of a vSwitch depends on the number of flows |
27 | | running through it. Also performance of other equipment or |
28 | | entities can depend on the number of flows or the packet |
31 | | The purpose is also to be able to spot the trends. |
32 | | Test results, graphs and similar shall be stored for |
33 | | comparison reasons and product evolution understanding |
34 | | between different OPNFV versions and/or configurations. |
36 +--------------+--------------------------------------------------------------+
39 | | Linux packet generator is a tool to generate packets at very |
40 | | high speed in the kernel. pktgen is mainly used to drive and |
41 | | LAN equipment test network. pktgen supports multi threading. |
42 | | To generate random MAC address, IP address, port number UDP |
43 | | packets, pktgen uses multiple CPU processors in the |
44 | | different PCI bus (PCI, PCIe bus) with Gigabit Ethernet |
45 | | tested (pktgen performance depends on the CPU processing |
46 | | speed, memory delay, PCI bus speed hardware parameters), |
47 | | Transmit data rate can be even larger than 10GBit/s. Visible |
48 | | can satisfy most card test requirements. |
50 | | (Pktgen is not always part of a Linux distribution, hence it |
51 | | needs to be installed. It is part of the Yardstick Docker |
53 | | As an example see the /yardstick/tools/ directory for how |
54 | | to generate a Linux image with pktgen included.) |
56 +--------------+--------------------------------------------------------------+
57 |test | This test case uses Pktgen to generate packet flow between |
58 |description | two hosts for simulating network workloads on the SUT. |
60 +--------------+--------------------------------------------------------------+
61 |traffic | An IP table is setup on server to monitor for received |
64 +--------------+--------------------------------------------------------------+
65 |configuration | file: opnfv_yardstick_tc001.yaml |
67 | | Packet size is set to 60 bytes. |
68 | | Number of ports: 10, 50, 100, 500 and 1000, where each |
69 | | runs for 20 seconds. The whole sequence is run twice |
70 | | The client and server are distributed on different hardware. |
72 | | For SLA max_ppm is set to 1000. The amount of configured |
73 | | ports map to between 110 up to 1001000 flows, respectively. |
75 +--------------+--------------------------------------------------------------+
76 |applicability | Test can be configured with different: |
79 | | * amount of flows; |
80 | | * test duration. |
82 | | Default values exist. |
84 | | SLA (optional): max_ppm: The number of packets per million |
85 | | packets sent that are acceptable to loose, not received. |
87 +--------------+--------------------------------------------------------------+
88 |usability | This test case is used for generating high network |
89 | | throughput to simulate certain workloads on the SUT. Hence |
90 | | it should work with other test cases. |
92 +--------------+--------------------------------------------------------------+
93 |references | pktgen_ |
97 +--------------+--------------------------------------------------------------+
98 |pre-test | The test case image needs to be installed into Glance |
99 |conditions | with pktgen included in it. |
101 | | No POD specific requirements have been identified. |
103 +--------------+--------------------------------------------------------------+
104 |test sequence | description and expected result |
106 +--------------+--------------------------------------------------------------+
107 |step 1 | Two host VMs are booted, as server and client. |
109 +--------------+--------------------------------------------------------------+
110 |step 2 | Yardstick is connected with the server VM by using ssh. |
111 | | 'pktgen_benchmark' bash script is copyied from Jump Host to |
112 | | the server VM via the ssh tunnel. |
114 +--------------+--------------------------------------------------------------+
115 |step 3 | An IP table is setup on server to monitor for received |
118 +--------------+--------------------------------------------------------------+
119 |step 4 | pktgen is invoked to generate packet flow between two server |
120 | | and client for simulating network workloads on the SUT. |
121 | | Results are processed and checked against the SLA. Logs are |
122 | | produced and stored. |
124 | | Result: Logs are stored. |
126 +--------------+--------------------------------------------------------------+
127 |step 5 | Two host VMs are deleted. |
129 +--------------+--------------------------------------------------------------+
130 |test verdict | Fails only if SLA is not passed, or if there is a test case |
131 | | execution problem. |
133 +--------------+--------------------------------------------------------------+