![]() ![]() The size of a bogo op depends on the stressor being run, and are not comparable between different stressors. Stress-ng measures a stress test "throughput" using "bogus operations per second". So we can see here that the mq stressor is forcing the processes to context switch at around 0.76 million per second, and we're getting quite low data cache misses. Stress-ng: info: 120.02s available CPU time Stress-ng: info: 0 Alignment Faults 0.00 sec Stress-ng: info: 952 CPU Migrations 31.73 sec Stress-ng: info: 0 Page Faults Major 0.00 sec ![]() Stress-ng: info: 176 Page Faults Minor 5.87 sec Stress-ng: info: successful run completed in 30.00s We can run the mq stressor with the -perf option to see some more detail on what the machine is doing during the run: Now consider a more interesting stress test, such as passing messages between processes using a POSIX message queue. The total CPU time was 4 x 60 seconds (240 seconds), of which 0.13% was in the kernel, and 85.50% in user time and stress-ng only got 85.64% of all the CPUs (since the machine was a bit busy doing other work at the same time). In the above example, I ran this on a machine that wasn't particularly idle with 4 CPU threads, so 4 instances were executed. Stress-ng: info: 240.00s available CPU time ![]() Stress-ng: info: successful run completed in 60.00s (1 min, 0.00 secs) Stress-ng: info: dispatching hogs: 4 matrix You can get an idea of how much user time and system (kernel) time is being used via the -times option: ![]() If you want to run an instance of this on ALL the CPUs on your machine, specify 0 instances and stress-ng will figure out how many to run: To run 1 instance of this for 60 seconds, use: Of all the tests, this one generally heats x86 CPUs the best. The matrix stressor is a good way to exercise the CPU floating point operations as well as memory and processor data cache. This document is a quick-start reference guide and covers some of the more typical use cases for stress-ng. The tool has a wide range of different stress mechanisms (known as "stressors") and a full description of these is included in the man page. Use stress-ng with caution as some of the tests can make a system run hot on poorly designed hardware and also can cause excessive system thrashing which may be difficult to stop. Stress-ng was originally intended to make a machine work hard and trip hardware issues such as thermal overruns as well as operating system bugs that only occur when a system is being thrashed hard. stress-ng also has a wide range of CPU specific stress tests that exercise floating point, integer, bit manipulation and control flow. It was designed to exercise various physical subsystems of a computer as well as the various operating system kernel interfaces. Stress-ng will stress test a computer system in various selectable ways. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |