Mathematica: ParallelMap getting slower

  • Context: Mathematica 
  • Thread starter Thread starter Polyrhythmic
  • Start date Start date
  • Tags Tags
    Mathematica
Click For Summary
SUMMARY

The discussion focuses on performance degradation observed when using ParallelMap in Mathematica on Linux for solving 100 eigenvalue problems across 4 kernels. Users report a significant drop in CPU usage for slave kernels, decreasing from 70% to approximately 20%, while the master kernel's CPU usage increases. The recommended solution includes utilizing the ClearSystemCache[] function at the end of processing loops to potentially mitigate this issue.

PREREQUISITES
  • Familiarity with Mathematica programming language
  • Understanding of parallel computing concepts
  • Knowledge of eigenvalue problem-solving techniques
  • Basic experience with Linux operating system
NEXT STEPS
  • Research the use of ClearSystemCache[] in Mathematica for performance optimization
  • Explore advanced parallel computing techniques in Mathematica
  • Investigate memory management strategies in Linux for computational tasks
  • Learn about profiling and monitoring CPU usage in Mathematica
USEFUL FOR

Mathematica users, computational scientists, and anyone involved in parallel processing and performance optimization in numerical computations.

Polyrhythmic
Messages
343
Reaction score
0
Hi,

I use ParallelMap with Mathematica on Linux to solve 100 eigenvalue problems on 4 kernels. This operation ist repeated very often, and over the course of time the performance drops significantly. Using top, you can see that the CPU used by the master kernel increases, and the CPU used by the slave kernels decreases from 70 to appr. 20 percent. Has anyone else experienced this phenomenon?
 
Physics news on Phys.org
Whats the associated ram usage? Try use ClearSystemCache[] at the end of your loops.
 

Similar threads

Mathematica Mathematica vs gnuplot vs matplotlib (efficency)
  • · Replies 1 ·
Replies
1
Views
9K
Questions about the operation of this C68MX11 CPU
  • · Replies 7 ·
Replies
7
Views
4K
Requesting suggestions for languages, libraries, and architectures for parallel (and sometimes non parallel) numerical and scientific computations
  • · Replies 8 ·
Replies
8
Views
5K
Thermal Dissipation of High-Power Electronic Subassemblies
  • · Replies 15 ·
Replies
15
Views
3K
Modern Digital Audio: Understanding Dither and Oversampling in Hi-Res Age
  • · Replies 17 ·
Replies
17
Views
6K
Graduate Vacuum Chamber Demo: Investigating Temperature & Humidity Changes
  • · Replies 3 ·
Replies
3
Views
3K
Mathematica This Week's Finds in Mathematical Physics (Week 242)
  • · Replies 1 ·
Replies
1
Views
4K
Mathematica This Week's Finds in Mathematical Physics (Week 228)
  • · Replies 2 ·
Replies
2
Views
3K
Mathematica This Week's Finds in Mathematical Physics (Week 228)
  • · Replies 2 ·
Replies
2
Views
4K
Challenge Micromass' big simulation challenge
  • · Replies 38 ·
2
Replies
38
Views
10K