1 *************************************
2 Yardstick Test Case Description TC012
3 *************************************
5 .. _man-pages: http://manpages.ubuntu.com/manpages/trusty/bw_mem.8.html
7 +-----------------------------------------------------------------------------+
10 +--------------+--------------------------------------------------------------+
11 |test case id | OPNFV_YARDSTICK_TC012_Memory Bandwidth |
13 +--------------+--------------------------------------------------------------+
14 |metric | Megabyte per second (MBps) |
16 +--------------+--------------------------------------------------------------+
17 |test purpose | Measure the rate at which data can be read from and written |
18 | | to the memory (this includes all levels of memory). |
20 +--------------+--------------------------------------------------------------+
21 |configuration | File: opnfv_yardstick_tc012.yaml |
23 | | * SLA (optional): 15000 (MBps) min_bw: The minimum amount of |
24 | | memory bandwidth that is accepted. |
25 | | * Size: 10 240 kB - test allocates twice that size (20 480kB)|
26 | | zeros it and then measures the time it takes to copy from |
27 | | one side to another. |
28 | | * Benchmark: rdwr - measures the time to read data into |
29 | | memory and then write data to the same location. |
30 | | * Warmup: 0 - the number of iterations to perform before |
31 | | taking actual measurements. |
32 | | * Iterations: 10 - test is run 10 times iteratively. |
33 | | * Interval: 1 - there is 1 second delay between each |
36 +--------------+--------------------------------------------------------------+
37 |test tool | Lmbench |
39 | | Lmbench is a suite of operating system microbenchmarks. This |
40 | | test uses bw_mem tool from that suite. |
41 | | Lmbench is not always part of a Linux distribution, hence it |
42 | | needs to be installed in the test image. |
44 +--------------+--------------------------------------------------------------+
45 |references | man-pages_ |
47 | | McVoy, Larry W., and Carl Staelin. "lmbench: Portable Tools |
48 | | for Performance Analysis." USENIX annual technical |
49 | | conference. 1996. |
51 +--------------+--------------------------------------------------------------+
52 |applicability | Test can be configured with different: |
55 | | * memory operations (such as rd, wr, rdwr, cp, frd, fwr, |
56 | | fcp, bzero, bcopy); |
57 | | * number of warmup iterations; |
58 | | * iterations and intervals. |
60 | | There are default values for each above-mentioned option. |
62 +--------------+--------------------------------------------------------------+
63 |pre-test | The test case image needs to be installed into Glance |
64 |conditions | with Lmbench included in the image. |
66 | | No POD specific requirements have been identified. |
68 +--------------+--------------------------------------------------------------+
69 |test sequence | description and expected result |
71 +--------------+--------------------------------------------------------------+
72 |step 1 | The host is installed as client. Lmbench's bw_mem tool is |
73 | | invoked and logs are produced and stored. |
75 | | Result: logs are stored. |
77 +--------------+--------------------------------------------------------------+
78 |test verdict | Test fails if the measured memory bandwidth is below the SLA |
79 | | value or if there is a test case execution problem. |
81 +--------------+--------------------------------------------------------------+