and pyperf also has docs on how to tune system for better benchmarking  Tune 
the system for benchmarks — pyperf 2.10.0 documentation 
<https://pyperf.readthedocs.io/en/latest/system.html>

On Monday, 9 March 2026 at 14:35:12 UTC+5:30 Krishna Chaitanya Bonu wrote:

> well here are some more articles I found regarding stable benchmarking.
>
> My journey to stable benchmark, part 1 (system) — Victor Stinner blog 3 
> <https://vstinner.github.io/journey-to-stable-benchmark-system.html>
> My journey to stable benchmark, part 2 (deadcode) — Victor Stinner blog 3 
> <https://vstinner.github.io/journey-to-stable-benchmark-deadcode.html>
> My journey to stable benchmark, part 3 (average) — Victor Stinner blog 3 
> <https://vstinner.github.io/journey-to-stable-benchmark-average.html>
>
> a command cpuset see here -> Cpuset (cset) Tutorial 
> <https://sources.debian.org/data/main/c/cpuset/1.6-4.1/doc/tutorial.html>. 
> I dont know much about cpuset and I haven't tried as I am not on a linux 
> machine.
>
> if someone has know it or have tried it please share info
> On Saturday, 7 March 2026 at 09:00:00 UTC+5:30 Krishna Chaitanya Bonu 
> wrote:
>
>> And also, why don't we use *pyperf*
>>
>> On Sunday, 1 March 2026 at 17:52:02 UTC+5:30 Krishna Chaitanya Bonu wrote:
>>
>>> hi everyone, i am new here...
>>>
>>> i looked at the benchmark project you have for gsoc...
>>> i have looked into this 
>>> https://pythonspeed.com/articles/consistent-benchmarking-in-ci/ and it 
>>> is quite interesting...
>>>
>>> I would like to know more about it...
>>> so If someone would send some articles about it or something related 
>>> that would really help...
>>>
>>> and using no of  instructions to benchmark looked crazy while reading 
>>> for the first time, but made a lot of sense.. and python has its own `dis` 
>>> module (would this module help us ???) to get instructions executed.
>>>
>>> one more question.
>>> if i remember correctly this project has been in the previous years in 
>>> gsoc as well. so i  would like to know what is done, and ofcourse what is 
>>> not done and the reasons behind the things that have not been done.
>>>
>>> thank you.
>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/sympy/9baaa412-04bc-4c59-af2f-a84d1e227230n%40googlegroups.com.

Reply via email to