diff options
Diffstat (limited to 'docs')
| -rw-r--r-- | docs/yardstick/opnfv_yardstick_tc012.rst | 68 |
1 files changed, 68 insertions, 0 deletions
diff --git a/docs/yardstick/opnfv_yardstick_tc012.rst b/docs/yardstick/opnfv_yardstick_tc012.rst new file mode 100644 index 000000000..0c8e9c02a --- /dev/null +++ b/docs/yardstick/opnfv_yardstick_tc012.rst @@ -0,0 +1,68 @@ + +************************************* +Yardstick Test Case Description TC012 +************************************* + ++-----------------------------------------------------------------------------+ +|Memory Bandwidth | ++==============+==============================================================+ +|test case id | OPNFV_YARDSTICK_TC012_Memory Bandwidth | ++--------------+--------------------------------------------------------------+ +|metric | Megabyte per second (MBps) | ++--------------+--------------------------------------------------------------+ +|test purpose | Measure the rate at which data can be read from and written | +| | to the memory (this includes all levels of memory). | ++--------------+--------------------------------------------------------------+ +|configuration | File: opnfv_yardstick_tc012.yaml | +| | | +| | * SLA (optional): 15000 (MBps) | +| | min_bw: The minimum amount of memory bandwidth that is | +| | accepted. | +| | * Size: 10 240 kB - test allocates twice that size (20 480kB)| +| | zeros it and then measures the time it takes to copy from | +| | one side to another. | +| | * Benchmark: rdwr - measures the time to read data into | +| | memory and then write data to the same location. | +| | * Warmup: 0 - the number of iterations to perform before | +| | taking actual measurements. | +| | * Iterations: 10 - test is run 10 times iteratively. | +| | * Interval: 1 - there is 1 second delay between each | +| | iteration. | ++--------------+--------------------------------------------------------------+ +|test tool | Lmbench | +| | | +| | Lmbench is a suite of operating system microbenchmarks. This | +| | test uses bw_mem tool from that suite. | +| | Lmbench is not always part of a Linux distribution, hence it | +| | needs to be installed in the test image | +| | (See :ref:`guest-image` for how to generate a Linux image | +| | for Glance with Lmbench included). | ++--------------+--------------------------------------------------------------+ +|references | * http://manpages.ubuntu.com/manpages/trusty/bw_mem.8.html | +| | | +| | * McVoy, Larry W., and Carl Staelin. "lmbench: Portable Tools| +| | for Performance Analysis." *USENIX annual technical | +| | conference*. 1996. | ++--------------+--------------------------------------------------------------+ +|applicability | Test can be configured with different | +| | * memory sizes; | +| | * memory operations (such as rd, wr, rdwr, cp, frd, fwr, | +| | fcp, bzero, bcopy); | +| | * number of warmup iterations; | +| | * iterations and intervals. | +| | There are default values for each above-mentioned option. | ++--------------+--------------------------------------------------------------+ +|pre-test | The test case image needs to be installed into Glance | +|conditions | with Lmbench included in the image. | +| | | +| | No POD specific requirements have been identified. | ++--------------+------+----------------------------------+--------------------+ +|test sequence | step | description | result | +| +------+----------------------------------+--------------------+ +| | 1 | The host is installed as client. | Logs are stored | +| | | Lmbench's bw_mem tool is invoked | | +| | | and logs are produced and stored.| | ++--------------+------+----------------------------------+--------------------+ +|test verdict | Test fails if the measured memory bandwidth is below the SLA | +| | value or if there is a test case execution problem. | ++--------------+--------------------------------------------------------------+ |