People here might have heard about the work of Prof Emery Berger on the causal profiling technology.
Here's a paper presenting Coz, a causal profiler.
https://www.sigops.org/s/conferences/sosp/2015/current/2015-Monterey/printable/090-curtsinger.pdf
They have also release a profiler for Python using the same technology.
It is called Scalene: https://github.com/plasma-umass/scalene
In case Pharo does not have such a profiler, maybe an expert could work on it?
Best,
~Mayuresh
Hello Mayuresh,
As you said, in Pharo we don't have an implementation of a casual profiler at the moments. I don't know if someone will implement one in a near future.
We have other profilers in Pharo: MessageTally and AndreasProfiler, both already in the image, are profilers that use sampling to get an estimated execution time for each method. Now we are working on a prototype of a memory profiler that we are going to release soon.
Regards,
Sebastian Jordan
----- Mail original -----
De: "mayuresh@kathe.in via Pharo-users" pharo-users@lists.pharo.org
À: "pharo-users" pharo-users@lists.pharo.org
Cc: mayuresh@kathe.in
Envoyé: Dimanche 15 Janvier 2023 07:28:57
Objet: [Pharo-users] Causal profiling
People here might have heard about the work of Prof Emery Berger on the causal
profiling technology.
Here's a paper presenting Coz, a causal profiler.
https://www.sigops.org/s/conferences/sosp/2015/current/2015-Monterey/printable/090-curtsinger.pdf
They have also release a profiler for Python using the same technology.
It is called Scalene: https://github.com/plasma-umass/scalene
In case Pharo does not have such a profiler, maybe an expert could work on it?
Best,
~Mayuresh