mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > Twin Prime Search

Reply
 
Thread Tools
Old 2010-06-07, 19:40   #12
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

31·67 Posts
Default

Okay, thanks. I'll try it with the 0.75T range I have left to do and see how much memory it uses.
Flatlander is offline   Reply With Quote
Old 2010-06-07, 19:53   #13
axn
 
axn's Avatar
 
Jun 2003

2×34×29 Posts
Default

Quote:
Originally Posted by henryzz View Post
It is possible to change the level of the recombine. Just set a limit on the sieve to shortly after you estimate 1/250(I would recommend 1/300 or less in case of mistakes) are remaining. That could in theory mean its possible to combine really early like 1e6 or something like that.
I will do a test now to see vaguely when.
edit: ~p=4e4 would do the trick nicely
How much time will this save? By my reckoning, you won't actually save any time at all.
axn is offline   Reply With Quote
Old 2010-06-07, 21:26   #14
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

81D16 Posts
Default

Quote:
Originally Posted by axn View Post
How much time will this save? By my reckoning, you won't actually save any time at all.
Possibly. Sieving a 0.05T range to 4e4 just took 27 mins, so < 10 hours for 1T. (Compare with c.24hrs to 1G.)
Sieving from 4e4 to 100G is progressing painfully slowly but of course the first bit is always much slower.

I think I'll just stop now and run a whole 1T range, now that I know it will easily fit in 485Mb once sieved to 4e4. I'll post the timings here later.

The NPG help file states:
"NewPGen is happy with lots of k's to sieve - there is nothing to be gained by dividing a range of k's up and sieving each subrange in turn..."
So I'm hoping there will be an increase in efficiency.

Last fiddled with by Flatlander on 2010-06-07 at 21:29
Flatlander is offline   Reply With Quote
Old 2010-06-07, 21:45   #15
axn
 
axn's Avatar
 
Jun 2003

111328 Posts
Default

Quote:
Originally Posted by Flatlander View Post
Sieving from 4e4 to 100G is progressing painfully slowly but of course the first bit is always much slower.
That's why I am asserting that there'll be no speed gain. In fact, due to the greater IO involved, it might actually be slower.
Quote:
Originally Posted by Flatlander View Post
The NPG help file states:
"NewPGen is happy with lots of k's to sieve - there is nothing to be gained by dividing a range of k's up and sieving each subrange in turn..."
So I'm hoping there will be an increase in efficiency.
Only true when p >= k range (or maybe range/2). Otherwise there is no increase in efficiency, and might even be slower due to memory pressure (Fast Array vs Normal Array mode).
axn is offline   Reply With Quote
Old 2010-06-07, 21:48   #16
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

31·67 Posts
Default

Fair enough.
Flatlander is offline   Reply With Quote
Old 2010-06-11, 22:25   #17
The Carnivore
 
The Carnivore's Avatar
 
Jun 2010

101111112 Posts
Default

Quote:
Originally Posted by Historian View Post
Processor: Pentium 4 3.4 GHz

tpsieve for the variable n-range: 5M p/sec
tpsieve for the variable n-range: 14.1M p/sec using one core of a 2.4 GHz core2duo processor.
The Carnivore is offline   Reply With Quote
Old 2010-06-11, 22:49   #18
amphoria
 
amphoria's Avatar
 
"Dave"
Sep 2005
UK

2×19×73 Posts
Default

Quote:
Originally Posted by axn View Post
Only true when p >= k range (or maybe range/2). Otherwise there is no increase in efficiency, and might even be slower due to memory pressure (Fast Array vs Normal Array mode).
This is in fact the problem with only sieving to p=4e4 in the first stage. It leaves about 32.5 M k's and therefore NewPGen uses normal array mode. I hadn't worked out why it was so slow until I stopped the sieve at p=1e6. When restarted NewPGen switched to fast array mode and the removal rate jumped from 40 k/sec to 280 k/sec, If there is an advantage (yet to be proven), then the first stage needs to sieve to much closer to p=1e6.
amphoria is offline   Reply With Quote
Old 2010-06-11, 23:22   #19
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

31×67 Posts
Default

Re. Megabit Twin Sieve
I tried sieving to 10M then again to 100G. The total time was 10-20% slower than letting NPG do it automatically.
I would seem that indeed the 'sweet spot' is above 1G.

Some scribbled timings:
Start to 10M took 15hrs 3m (Leaving 6.1M ks.)
Started NPG again. (Used fast array, 384Mb ram.)
10M-20M took 1hr 24m
20m-100m took 4hrs 25m
100m-500m took 2hrs 41m
500m-100G to 8hrs 33m

Total time c. 32hrs compared with c. 26-28 hrs (iirc) letting NPG do it automatically.

C2Quad Q6700 at stock 2.66GHz. 2Gb ram, NPG memory at maximum 485Mb.

Last fiddled with by Flatlander on 2010-06-11 at 23:27 Reason: Clarification.
Flatlander is offline   Reply With Quote
Old 2010-06-11, 23:34   #20
axn
 
axn's Avatar
 
Jun 2003

111328 Posts
Default

Quote:
Originally Posted by amphoria View Post
This is in fact the problem with only sieving to p=4e4 in the first stage. It leaves about 32.5 M k's and therefore NewPGen uses normal array mode. I hadn't worked out why it was so slow until I stopped the sieve at p=1e6. When restarted NewPGen switched to fast array mode and the removal rate jumped from 40 k/sec to 280 k/sec, If there is an advantage (yet to be proven), then the first stage needs to sieve to much closer to p=1e6.
Hmmm... How much time did the sieve take to get to 4e4? 1e6? 10e6?

I am asking because I am "fairly confident" that I can write a custom sieve that can sieve a range of k's to 1e6 (or even 10e6) much faster than NewPGen can. NewPGen isn't really optimised for the initial sieving.
axn is offline   Reply With Quote
Old 2010-06-11, 23:35   #21
axn
 
axn's Avatar
 
Jun 2003

10010010110102 Posts
Default

Quote:
Originally Posted by Flatlander View Post
Some scribbled timings:
Start to 10M took 15hrs 3m (Leaving 6.1M ks.)
Hmmm. I should be able to do better than this. Gotta code it up, though, to know for sure.
axn is offline   Reply With Quote
Old 2010-06-11, 23:48   #22
Flatlander
I quite division it
 
Flatlander's Avatar
 
"Chris"
Feb 2005
England

207710 Posts
Default

Quote:
Originally Posted by axn View Post
Hmmm... How much time did the sieve take to get to 4e4? 1e6? 10e6?...
Quote:
Originally Posted by Flatlander View Post
... Sieving a 0.05T range to 4e4 just took 27 mins, so < 10 hours for 1T...
All I have is '7hrs 46m to 4e4' jotted on a piece of paper but that sounds a bit quick.

All figures subject to slight variation as NPG had high priority while 4 x LLRNet with low priority were running in the background. (So I didn't waste any cycles when NPG finished overnight.)
Flatlander is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Perpetual benchmark thread... Xyzzy Hardware 821 2020-09-12 23:56
Hardware Benchmark Jest Thread for 100M exponents joblack Hardware 275 2019-08-04 21:07
LLR benchmark thread Oddball Riesel Prime Search 5 2010-08-02 00:11
sr5sieve Benchmark thread axn Sierpinski/Riesel Base 5 25 2010-05-28 23:57
New Sieve Thread Discussion Citrix Prime Sierpinski Project 15 2005-08-29 13:56

All times are UTC. The time now is 04:13.

Tue Sep 29 04:13:20 UTC 2020 up 19 days, 1:24, 0 users, load averages: 1.98, 1.91, 1.77

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.