mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > PrimeNet

Reply
 
Thread Tools
Old 2020-01-06, 03:23   #1
drew
 
drew's Avatar
 
Jun 2005

3·127 Posts
Default Worker Windows setting

Hi,

For years, I've set my worker window settings to use Prime95 single-threaded by using 4 worker windows (1 per physical CPU). This optimizes the throughput, as far as I can tell.

For the first time I looked at the GIMPS assignment rules closely and realized these rules are set up in a way that prioritizes completing assignments quickly, giving faster workers smaller exponents.

This has the odd effect that by splitting up the work, which increases the duration of each test, I'll get assigned larger exponents, thereby taking even longer.

Should I be allocating the worker windows differently with the goal of completing smaller exponents sooner rather than optimizing the overall throughput?
drew is offline   Reply With Quote
Old 2020-01-06, 04:16   #2
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

26×67 Posts
Default

Quote:
Originally Posted by drew View Post
Hi,

For years, I've set my worker window settings to use Prime95 single-threaded by using 4 worker windows (1 per physical CPU). This optimizes the throughput, as far as I can tell.

For the first time I looked at the GIMPS assignment rules closely and realized these rules are set up in a way that prioritizes completing assignments quickly, giving faster workers smaller exponents.

This has the odd effect that by splitting up the work, which increases the duration of each test, I'll get assigned larger exponents, thereby taking even longer.

Should I be allocating the worker windows differently with the goal of completing smaller exponents sooner rather than optimizing the overall throughput?
My experience would suggest that changing to 2 workers with 2 cores each "should not" reduce overall thruput. That is, if running 4 workers each assignment takes 30 days then with 2 workers with 2 cores each assignment should be close to 15 days; maybe 16 days.

However extending to 1 worker with 4 cores will likely reduce overall thruput; I'll hazard a guess of 20 or 30%. Each assignment would not drop to 7.5 or 8 days but more likely 10. It depends a lot on amount or RAM; speed of RAM and the internal BUS/Cache etc. hocus-pocus (that means I'm a novice on this stuff).

That all said your best friend is simply to run a benchmark and it will tell you exactly.
petrw1 is offline   Reply With Quote
Old 2020-01-06, 04:31   #3
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

11·347 Posts
Default

Don't guess. Benchmark. It depends on the processor, cache size, and the fft size, whether 1, 2, 4, etc cores/worker are maximal throughput overall. See the attachments at posts 4 and 5 of https://www.mersenneforum.org/showthread.php?t=23900
As exponents get larger, primality tests get longer, and more cores/worker may be needed to avoid expiration before assignment completion. This is especially the case with older slower multicore processors.

Last fiddled with by kriesel on 2020-01-06 at 04:33
kriesel is offline   Reply With Quote
Old 2020-01-06, 04:55   #4
Uncwilly
6809 > 6502
 
Uncwilly's Avatar
 
"""""""""""""""""""
Aug 2003
101×103 Posts

25×251 Posts
Default

The push for faster turn over with regards to the assignment rules is primarily about the trailing edge. If you are in Cat 2 or 3, you should not be having a problem finishing the exponents using 1 core per worker. Cat 0 and 1 will require 2 cores. (Provided you keep your Days of work to queue low.)
Uncwilly is offline   Reply With Quote
Old 2020-01-06, 06:30   #5
drew
 
drew's Avatar
 
Jun 2005

3·127 Posts
Default

Quote:
Originally Posted by Uncwilly View Post
The push for faster turn over with regards to the assignment rules is primarily about the trailing edge. If you are in Cat 2 or 3, you should not be having a problem finishing the exponents using 1 core per worker. Cat 0 and 1 will require 2 cores. (Provided you keep your Days of work to queue low.)
Ok thanks. I get the point about trying to capture the trailing edge. I'm not worried about being able to complete the work in time, I was just trying to understand what's in the best interest for GIMPS.
drew is offline   Reply With Quote
Old 2020-01-06, 08:16   #6
kladner
 
kladner's Avatar
 
"Kieren"
Jul 2011
In My Own Galaxy!

2×32×547 Posts
Default

Quote:
Originally Posted by petrw1 View Post
My experience would suggest that changing to 2 workers with 2 cores each "should not" reduce overall thruput. That is, if running 4 workers each assignment takes 30 days then with 2 workers with 2 cores each assignment should be close to 15 days; maybe 16 days.

However extending to 1 worker with 4 cores will likely reduce overall thruput; I'll hazard a guess of 20 or 30%. Each assignment would not drop to 7.5 or 8 days but more likely 10. It depends a lot on amount or RAM; speed of RAM and the internal BUS/Cache etc. hocus-pocus (that means I'm a novice on this stuff).

That all said your best friend is simply to run a benchmark and it will tell you exactly.
I'm not sure this is true. Would contrary benchmarks from a one worker, four core setup help cast light on the subject?

"Scuze me while I whip this out!"

(See attached.) I just stopped my normal DC work and ran this short-form benchmark. The CPU frequency of 4000MHz given is incorrect. The CPU runs 4 cores locked at 4300MHz. The BIOS shows this (4000MHz) at startup, as well. But a short test run of the benchmark with HWINFO running showed the CPU at 4300. The same is true with regular DCLL work.
Attached Files
File Type: txt bench01.txt (5.6 KB, 28 views)

Last fiddled with by kladner on 2020-01-06 at 18:00
kladner is offline   Reply With Quote
Old 2020-01-06, 08:30   #7
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

11×347 Posts
Default

Quote:
Originally Posted by Uncwilly View Post
The push for faster turn over with regards to the assignment rules is primarily about the trailing edge. If you are in Cat 2 or 3, you should not be having a problem finishing the exponents using 1 core per worker. Cat 0 and 1 will require 2 cores. (Provided you keep your Days of work to queue low.)
It depends a great deal on how slow the cores are.
kriesel is offline   Reply With Quote
Old 2020-01-06, 15:51   #8
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

428810 Posts
Default

Quote:
Originally Posted by kladner View Post
I'm not sure this is true. Would contrary benchmarks from a one worker, four core setup help cast light on the subject?

"Scuze me while I whip this out!"

(See attached.) I just stopped my normal DC work and ran this short-form benchmark. The CPU frequency of 4000MHz given is incorrect. The CPU runs 4 cores locked at 4300MHz. The BIOS shows this at startup, as well. But a short test run of the benchmark with HWINFO running showed the CPU at 4300. The same is true with regular DCLL work.
Hmmm as I read it I stand corrected.
Geez usually my 1 mistake of the year takes longer. LOL
petrw1 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
CPU Worker Windows storm5510 Software 10 2017-06-23 06:34
How to reduce number of worker windows? Chuck PrimeNet 7 2011-07-03 19:17
worker windows: one per physical/logical core? ixfd64 Software 2 2010-12-09 17:38
Worker Windows - Optimal settings Unregistered Information & Answers 4 2010-07-30 21:49
Worker Windows question joblack Software 1 2009-01-02 00:24

All times are UTC. The time now is 11:23.

Sat May 30 11:23:50 UTC 2020 up 66 days, 8:56, 2 users, load averages: 2.35, 1.81, 1.58

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.