Julia not using full CPU when running inside a GNU Screen session (edit: not a Julia isse


#1

EDIT: the following is not a Julia issue, I notice exactly the same problem using Numpy.

The following code:

x=rand(10000,10000);
x*x

shows ~800% CPU usage according to top (I have 8 cores). When running Julia within a GNU Screen session however, I see only ~100% usage. I am running on Ubuntu 16.04, installed Julia by downloading from the web page and my versioninfo() is

Julia Version 0.6.2
Commit d386e40c17 (2017-12-13 18:08 UTC)
Platform Info:
  OS: Linux (x86_64-pc-linux-gnu)
  CPU: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
  WORD_SIZE: 64
  BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Sandybridge)
  LAPACK: libopenblas64_
  LIBM: libopenlibm
  LLVM: libLLVM-3.9.1 (ORCJIT, sandybridge)

Screen:

Screen version 4.03.01 (GNU) 28-Jun-15

#2

Axel, stupid question. How exactly do you know the amount of cores in use?

I am sure I am going to be laughed at, but is screen somehow being put into a cgroup on your system?


#3

I didn’t: I just see that top shows ~100% CPU usage when running inside screen, while 800% (I have 8 cores) is used when running without screen. Thanks for pointing this out, I don’t have any proof that only 1 core is used.


#4

htop will you show you individual core usage.