mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Hardware > GPU Computing > GpuOwl

Reply
 
Thread Tools
Old 2020-10-16, 04:53   #1
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

116916 Posts
Default Relative performance of GPUs for P1

I realize P1 as a separate task is discontinued ... however ...

I am still running the version that allows it:
Does it seems reasonable that for the various Colab GPUs available I am seeing relative Stage1 iteration times of (based on my specific B1 but still relative):

P4: 3,600
T4: 2,630
K80: 1,800
P100: 470 (yes 4 to 8 times faster)
petrw1 is offline   Reply With Quote
Old 2020-10-16, 12:15   #2
tServo
 
tServo's Avatar
 
"Marv"
May 2009
near the Tannhäuser Gate

2·271 Posts
Default

Quote:
Originally Posted by petrw1 View Post
I realize P1 as a separate task is discontinued ... however ...

I am still running the version that allows it:
Does it seems reasonable that for the various Colab GPUs available I am seeing relative Stage1 iteration times of (based on my specific B1 but still relative):

P4: 3,600
T4: 2,630
K80: 1,800
P100: 470 (yes 4 to 8 times faster)
Yes, these times make perfect sense.

Neither the P4 nor the T4 have many FP64 cores available. These cores are essential for performance doing Stage1. Their specs are fairly close but since the T4 is newer with faster memory & a few other things, it should be faster than the P4.
Even tho the K80 is quite old, it still has decent FP64 performance AND it has 2 GPUs.
The P100 has lots of FP64 cores and they will yield the best performance.

AFAIK the P4 and T4 are touted as being designed explicitly for training AIs since they do not require high percision computations.

Last fiddled with by tServo on 2020-10-16 at 12:16
tServo is offline   Reply With Quote
Old 2020-10-16, 13:46   #3
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

221448 Posts
Default

Quote:
Originally Posted by tServo View Post
Even tho the K80 is quite old, it still has decent FP64 performance AND it has 2 GPUs.
The K80 itself has two (2#) GPUs on the card, but only one (1#) is given to each VM.
chalsall is online now   Reply With Quote
Old 2020-10-16, 14:12   #4
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

37·127 Posts
Default

Quote:
Originally Posted by petrw1 View Post
I realize P1 as a separate task is discontinued ... however ...

I am still running the version that allows it:
Does it seems reasonable that for the various Colab GPUs available I am seeing relative Stage1 iteration times of (based on my specific B1 but still relative):

P4: 3,600
T4: 2,630
K80: 1,800
P100: 470 (yes 4 to 8 times faster)
us/iteration for ~100M exponents? Time required for any fft-based multiplication mod m is strongly related to log2(m); roughly p1.1 for Mersenne number m=2p-1. Some data for Colab gpus at https://www.mersenneforum.org/showpo...5&postcount=15, showing the P4 & T4 have 1/32 SP/DP ratio, making them better suited for TF, not well suited for LL, PRP, P-1.
kriesel is offline   Reply With Quote
Old 2020-10-16, 18:19   #5
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

116916 Posts
Default

Quote:
Originally Posted by kriesel View Post
us/iteration for ~100M exponents?
44.6M
B1=1,250,000
B2=25,000,000
petrw1 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Gaps in k-almost primes relative to 2^k, a question mart_r Prime Gap Searches 3 2020-04-02 19:36
was this formula known relative to primality and factorization that you know? Alberico Lepore Alberico Lepore 7 2019-01-03 23:42
Number of relative primes for P-1 petrw1 Factoring 4 2018-11-07 20:20
P-1 factoring, relative primes timbit Information & Answers 0 2009-03-13 18:45
relative speed of processors Primeinator Hardware 10 2005-02-27 18:03

All times are UTC. The time now is 16:57.

Mon Nov 23 16:57:31 UTC 2020 up 74 days, 14:08, 2 users, load averages: 2.74, 2.01, 1.84

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.