Project

General

Profile

Feature #1417

Feature request: Parallel running

Added by Dobbs, Adam over 10 years ago. Updated over 10 years ago.

Status:
Open
Priority:
Normal
Category:
Code Management
Target version:
Start date:
18 February 2014
Due date:
% Done:

0%

Estimated time:
Workflow:
New Issue

Description

More commonly now people are starting to use MAUS for larger scale analysis runs, such as 10k - 100k event runs, which take a good number of hours. The architecture is built to support parallel processing, but this is not at present implemented. It would be a great help if we could exploit multiple cores with MAUS (not just on the grid batch production version).


Files

distributed_offline.py (12.5 KB) distributed_offline.py modified version of analyze_data_online for parallel offline running Rajaram, Durga, 05 March 2014 10:48
reduce_output.py (910 Bytes) reduce_output.py outputter Rajaram, Durga, 05 March 2014 10:48
sh_distributed.png (7.24 KB) sh_distributed.png #slab hits from distributed_offline Rajaram, Durga, 06 March 2014 08:31
sh_distributed-run2.png (7.08 KB) sh_distributed-run2.png #slab hits from distributed_offline -- rerun Rajaram, Durga, 06 March 2014 08:31
distributed_offline_logs.tar.gz (383 KB) distributed_offline_logs.tar.gz logs from running distributed_offline Rajaram, Durga, 06 March 2014 08:31
nspills_offline.png (5.33 KB) nspills_offline.png plot of spill-number from analyze_data_offline Rajaram, Durga, 06 March 2014 08:31
nspills_distributed.png (5.6 KB) nspills_distributed.png plot of spill-number from distributed_offline Rajaram, Durga, 06 March 2014 08:31
nspills_distributed-run2.png (5.52 KB) nspills_distributed-run2.png plot of spill-number from distributed_offline -- rerun Rajaram, Durga, 06 March 2014 08:31
sh_offline.png (7.49 KB) sh_offline.png #slab hits from analyze_data_offline Rajaram, Durga, 06 March 2014 08:31

Also available in: Atom PDF