mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   No Prime Left Behind (https://www.mersenneforum.org/forumdisplay.php?f=82)
-   -   Sieving drive for k=1003-2000 n=500K-1M (https://www.mersenneforum.org/showthread.php?t=11487)

gd_barnes 2009-02-12 04:34

Sieving drive for k=1003-2000 n=500K-1M
 
[COLOR=black][FONT=Verdana]This is a sieving drive for k=[/FONT][/COLOR][COLOR=black][FONT=Verdana]1003-2000 and n=500K-1M. I'll consider this to be "moderate" priority at this point but I would be grateful for any help that people can give. The 8th drive is progressing much faster than expected. Although not a requirement, it would be nice if we had the range sieved enough so that n=500K-600K was ready at about the time that the 8th drive is complete.[/FONT][/COLOR]

I started the sieve on 2 cores and later 1 quad several weeks ago and have now reached P=1T. I now have 3 quads on it and they are sieving P=1000G-4000G with expected completion in ~15 days. More specifically, each core is sieveing P=250G and the ranges are running on 2.4-GHZ Intel Quads on 64-bit Linux Ubuntu running 64-bit sr2sieve.

Because the 8th drive has progressed so quickly, I think the best thing to do for this effort is to sieve until we reach an optimum depth for just the lower portion of n=500K-600K, break it off for LLRing, and continue sieving n=600K-1M so that we don't have much down time on this k-range. Although not calculated yet, I'm estimating that the optimum for n=500K-600K is P=~15T and for n=600K-1M is P=~30T. We'll be able to compute a more exact optimum depth once we complete sieving to P=4T (4000G).

As you probably know, the sieve (.dat) file is huge. Even in ABCD format, the entire range sieved to P=1T is 25MB in full and 8MB zipped. I have now Emailed the sieve file to everyone.

Please take no more than ~2 weeks of work to start with. As I implied previously, on a modern machine with a 64-bit O.S. and 64-bit sr2sieve, it takes one core ~15 days to sieve a P=250G range.

Like before, sr2sieve is what we will use. Let us know if you need the executable or more detailed instructions on using it.

I prefer that we use the more "modern" method of sieving by specifying the ranges and file name at the command prompt as in:

sr2sieve -p 4000e9 -P 4250e9 -i sieve1003-2000-500K-1M.txt

The above would be if you were sieving P=4000G-4250G. The file is listed after the "-i" command and is the actual file name that will be in the zipped file at googlepages. Feel free to name it something shorter if you want or use the "srwork" older convention where you don't have to specify a file name.

When complete, you should have a factors.txt file. Just post the file here in this thread or if it is too big, please Email the file to me at:
gbarnes017 at gmail dot com

Note: The sieve file has been Emailed to all current sievers. If you are new to the effort and would like to help sieve, please post your reservation here and I'll Email you the file. All factors have now been removed up to P=20T. As we progress, I'll continue removing factors for contiguous P-ranges.

Reservations:
[code]
P-range reserved by status est. completion date
0G-4000G gd_barnes complete
4000G-4250G MyDogBuster complete
4250G-4500G PCZ complete
4500G-4750G MyDogBuster complete
4750G-5000G PCZ complete
5000G-5250G MyDogBuster complete
5250G-5500G PCZ complete
5500G-5750G MyDogBuster complete
5750G-6000G PCZ complete
6000G-6250G Flatlander complete
6250G-9250G gd_barnes complete
9250G-10000G PCZ complete
10000G-10250G Flatlander complete
10250G-11000G MyDogBuster complete
11000G-12500G PCZ complete
12500G-15000G gd_barnes complete
15000G-16000G MyDogBuster complete
16000G-17000G PCZ complete
17000G-17250G Flatlander complete
17250G-19750G Brucifer complete
19750G-19850G gd_barnes complete
19850G-20000G Brucifer complete
20000G-25000G Brucifer complete
25000G-25500G Flatlander complete
25500G-28300G Brucifer complete
28300G-29750G Brucifer complete
29750G-30000G MyDogBuster complete
30000G-40000G IronBits complete
40000G-45000G PCZ complete
No more sieving will be needed.

[COLOR=Red][B]Nice job all.[/B][/COLOR]

[/code]Once this is done, we'll have well-sieved files for the entire range of k=300-2000 up to n=1M! :smile:


Gary

Flatlander 2009-02-12 13:11

[quote]I prefer that we use the more "modern" method of sieving by specifying the ranges and file name at the command prompt as in:

sr2sieve -p 4000e9 -P 4250e9 -i sieve1003-2000-500K-1M.txt[/quote]Will this method save checkpoint files so that the sieve can be restarted with just "sr2sieve" or is it necessary* to use the other method for that?

*(I'm 44 and I [I]still[/I] have to check the spelling each time!)

Mini-Geek 2009-02-12 13:41

[quote=Flatlander;162571]Will this method save checkpoint files so that the sieve can be restarted with just "sr2sieve" or is it necessary* to use the other method for that?

*(I'm 44 and I [I]still[/I] have to check the spelling each time!)[/quote]
Yes, this method will save checkpoint files and when restarted with the same command will ignore your starting p and use the last one from the checkpoint. If I'm wrong (which I'm almost certain I'm not since I've done this before), please correct me. :smile:

mdettweiler 2009-02-12 15:22

[quote=Mini-Geek;162573]Yes, this method will save checkpoint files and when restarted with the same command will ignore your starting p and use the last one from the checkpoint. If I'm wrong (which I'm almost certain I'm not since I've done this before), please correct me. :smile:[/quote]
As Mini-Geek said--just restart it with the same command and it will pick up where it left off. I usually use a batch file (or, on Linux, a shell script) to start sr2sieve each time so that I don't have to type that long command each time. Alternatively, you could use an "sr2sieve-command-line.txt" file, which would serve a similar purpose as a batch file/shell script, but just through different means.

henryzz 2009-02-12 16:22

[quote=Flatlander;162571]Will this method save checkpoint files so that the sieve can be restarted with just "sr2sieve" or is it necessary* to use the other method for that?

*(I'm 44 and I [I]still[/I] have to check the spelling each time!)[/quote]
do you use firefox? there is a brilliant add-on that checks spelling for everything you type and puts red zigzag lines underneath incorrect spellings.
maybe that would help.:smile:

Mini-Geek 2009-02-12 16:57

[quote=henryzz;162584]do you use firefox? there is a brilliant add-on that checks spelling for everything you type and puts red zigzag lines underneath incorrect spellings.
maybe that would help.:smile:[/quote]
In the newest version of Firefox it's built-in, but is actually a dotted line, not a zigzag.

MyDogBuster 2009-02-12 17:02

Gary, I did a test run om my old AMD Quad. Doing 500G per core gives me an estimate of Apr 26 to do 2T. Does this sound about right. Thats 74 days to do 2T on 1 machine. Your doing 1T in 15 days. Using SR2Sieve 1.7.15, the same program I used for the last sieve effort.

OBTW, Reserving 4000G to 6000G

Flatlander 2009-02-12 17:23

[quote=henryzz;162584]do you use firefox? there is a brilliant add-on that checks spelling for everything you type and puts red zigzag lines underneath incorrect spellings.
maybe that would help.:smile:[/quote]
Thanks.
I have 'check my spelling as I type' checked but it does nothing.
What's the add-on called?

henryzz 2009-02-12 20:15

[quote=Flatlander;162598]Thanks.
I have 'check my spelling as I type' checked but it does nothing.
What's the add-on called?[/quote]
[url]https://addons.mozilla.org/en-US/firefox/addon/3366[/url]

Mini-Geek 2009-02-12 20:41

If you want to try to get the built-in function working, try installing this British English dictionary and make sure it's selected under Languages (just under Check Spelling) in the context menu of a text area:
[URL]https://addons.mozilla.org/en-US/firefox/addon/3366[/URL]
Edit: I just noticed that this is the same thing henryzz linked to. It's not really an extension on its own, just a dictionary to make Firefox's built in thing work right for Brits.

Flatlander 2009-02-12 21:01

Working now. Thanks! :smile:

gd_barnes 2009-02-13 22:43

[quote=MyDogBuster;162590]Gary, I did a test run om my old AMD Quad. Doing 500G per core gives me an estimate of Apr 26 to do 2T. Does this sound about right. Thats 74 days to do 2T on 1 machine. Your doing 1T in 15 days. Using SR2Sieve 1.7.15, the same program I used for the last sieve effort.

OBTW, Reserving 4000G to 6000G[/quote]


No. I'm doing P=250G per core in 15-16 days (i.e. 1T per quad). You're doing P=500G per core in 74 days (i.e 2T on a quad). If it was the same speed as mine, it shoudl take you 30-32 days per core. Hold old is your quad. Although I'm using 2.4 Ghz Intel's, the AMD's should be nearly as fast for sieving. Are you running 64-bit sr2sieve on a 64-bit O.S.? Since it's an older AMD, I would almost exactly twice that speed, i.e. P=500G per core in 37 days.

Are you reserving 74 days work? If so, I'd prefer if people only reserve about 2 weeks worth of work at a time. I'm requesting that in case we have a "heavy player" come along who swoops up and is able to complete everything to P=~15T in 2-3 weeks. We don't straggling ranges holding us up at that point.

On another note, Chris, are you reserving any sieving work? I couldn't tell by your long commentary here.


Gary

MyDogBuster 2009-02-13 22:58

[quote]Are you reserving 74 days work? If so, I'd prefer if people only reserve about 2 weeks worth of work at a time. I'm requesting that in case we have a "heavy player" come along who swoops up and is able to complete everything to P=~15T in 2-3 weeks. We don't straggling ranges holding us up at that point.
[/quote]

This AMD quad has always been about twice as slow as my Intels. I even added 2gb (up to 5gb) of memory yesterday and it sped it up about about 5%. I'm trying to find a Bios replacement. It has to be the mobo thats holding it back.

I've already started the 74 days of work. Besides, this is the only machine I have available. After this batch, I'll reserve just 2 weeks worth. This machine would been ideal for that other sieveing effort we talked about a month ago.

PS: I checked, I have the 64bit SR2Sieve on a 64 bit machine.

gd_barnes 2009-02-13 23:07

[quote=MyDogBuster;162748]This AMD quad has always been about twice as slow as my Intels. I even added 2gb (up to 5gb) of memory yesterday and it sped it up about about 5%. I'm trying to find a Bios replacement. It has to be the mobo thats holding it back.

I've already started the 74 days of work. Besides, this is the only machine I have available. After this batch, I'll reserve just 2 weeks worth. This machine would been ideal for that other sieveing effort we talked about a month ago.

PS: I checked, I have the 64bit SR2Sieve on a 64 bit machine.[/quote]


OK, but it's on your hide if we're nearing the optimum sieve depth for n=500K-600K within about 45 days. (lol) With me reseving P=3T every 15-16 days, we'll progress fairly quickly.

Flatlander 2009-02-13 23:15

[quote=gd_barnes;162742]...
On another note, Chris, are you reserving any sieving work? I couldn't tell by [B]your long commentary here[/B].
Gary[/quote]

Oh, the irony! :grin:

I have a stability issue with Adobe Premiere Elements a.t.m. (finally treated myself :smile:) but should be able to reserve a range soon.

MyDogBuster 2009-02-13 23:36

[QUOTE]OK, but it's on your hide if we're nearing the optimum sieve depth for n=500K-600K within about 45 days. (lol) With me reseving P=3T every 15-16 days, we'll progress fairly quickly. [/QUOTE]

Well if that happens, then they can just concentrate on Drives 5,6,and 7 till the Advanced Molasses Device catches up.

gd_barnes 2009-02-15 12:34

[quote=Flatlander;162753]Oh, the irony! :grin:
[/quote]

Ha-ha-ha! Good one! BTW, what the world are pound shops? Are those like pawn shops like we have in the U.S. where people that need money can sell their stuff?

[quote=MyDogBuster;162754]Well if that happens, then they can just concentrate on Drives 5,6,and 7 till the Advanced Molasses Device catches up.[/quote]

Just as good a one! lol

Mini-Geek 2009-02-15 13:11

[quote=gd_barnes;162871]BTW, what the world are pound shops? Are those like pawn shops like we have in the U.S. where people that need money can sell their stuff?[/quote]
[QUOTE][B]pound shop[/B]
[I]noun[/I] [LIST=1][*](UK / Ireland) a shop selling household groceries, toys, gimmicks etc, that are typically priced at a pound each.[/LIST][/QUOTE]Considering the UK currency is called "pound", I equated it to a "dollar store", which is correct according to this definition. (but, obviously, priced at a pound instead of at a dollar)

Flatlander 2009-02-16 15:46

[FONT=monospace]Sieving 6000G-6250G.

Running on Windows 7 beta 64 bit @ 2.38GHz.

ETA 3th-4th March.:wink:

[/FONT][URL]http://www.poundland.com/pages/default.aspx[/URL]
:grin:
[FONT=monospace]



[/FONT]

gd_barnes 2009-02-20 05:37

[quote=MyDogBuster;162754]Well if that happens, then they can just concentrate on Drives 5,6,and 7 till the Advanced Molasses Device catches up.[/quote]


Ian,

Could I talk you into halving your ranges on each core while I'll pick up the remainder of each starting on Feb. 26th or so?

I know it sounds a little messy but it shouldn't be. Just let me know what ranges you won't be doing and I'll load one of my 3 quads up with those ranges after my current P=1T-4T range is done on the above date. Assuming that we do that, I'll reserve only a P=2T range instead for the other 2 quads. Alternatively, I may add a 4th quad to the effort. [If you do that, I'm assuming that you would give up P=4250G-4500G, 4750G-5000G, 5250G-5500G, & 5750G-6000G and will only process P=4000G-4250G, 4500G-4750G, 5000G-5250G, & 5500G-5750G.]

All you'll have to do is a Ctl-C to stop each and restart each with a new maximum P-range. Obviously the minimum range would remain the same. Sr2sieve will just start where it left off at.

By doing that, I think there will be a very good chance that we'll be close to ready by the end of March to break off n=500K-600K. I hate to stop the very prolific and popular k=1400-2000 range for too long.


Thanks,
Gary

MyDogBuster 2009-02-20 07:51

[QUOTE]I know it sounds a little messy but it shouldn't be. Just let me know what ranges you won't be doing and I'll load one of my 3 quads up with those ranges after my current P=1T-4T range is done on the above date. Assuming that we do that, I'll reserve only a P=2T range instead for the other 2 quads. Alternatively, I may add a 4th quad to the effort. [If you do that, I'm assuming that you would give up P=4250G-4500G, 4750G-5000G, 5250G-5500G, & 5750G-6000G and will only process P=4000G-4250G, 4500G-4750G, 5000G-5250G, & 5500G-5750G.][/QUOTE]

Done

I've restarted with half the ranges as stated by you above.

gd_barnes 2009-02-20 08:25

[quote=MyDogBuster;163345]Done

I've restarted with half the ranges as stated by you above.[/quote]

Great!

Brian has responded favorably to me requesting that he throw a quad at the sieving effort. I'll just suggest that he do these ranges.

Thanks!


Gary

PCZ 2009-02-20 09:03

I will pick up the following ranges
4250G-4500G
4750G-5000G
5250G-5500G
5750G-6000G

No ETA yet

gd_barnes 2009-02-24 10:42

Ian, Brian, and Chris, can you post an ETA on your sieve ranges? I'm going to put them in the 1st post of this thread. It'll help me determine if I need to put more and or can put less cores on the effort.

My current range will complete on Feb. 26th.


Thanks,
Gary

MyDogBuster 2009-02-24 11:30

19th of Mar for me.

Flatlander 2009-02-24 14:14

ETA 3rd March.

(I'm still recovery from the disappointment of [I]not [/I]finding 35 primes in one day.)

PCZ 2009-02-25 02:29

5th Mar

gd_barnes 2009-02-25 05:10

Thanks guys. I noted the ETA's in the 1st post of this thread.

Sometime on Feb. 26th, I'll reserve another P=3T range. Likely on March 14th-15th, I'll get another P=3T range. I may reserve a little more on the 14th-15th reservation depending on what others have reserved.

I'm hoping that we can make P=15T by March 31st. Regardless, that is the day that we'll break off n=500K-600K. If the 8th drive is done at that point, I'm taking Chris's advice and starting the 10th drive with that smaller n-range, even if we aren't quite sieved far enough. P=15T would be 3 times further sieved than the same n-range for the 1st drive so that's sufficient as far as I'm concerned.


Gary

mdettweiler 2009-02-25 05:14

[quote=gd_barnes;163926]I'm hoping that we can make P=15T by March 31st. Regardless, that is the day that we'll break off n=500K-600K. If the 8th drive is done at that point, I'm taking Chris's advice and starting the 10th drive with that smaller n-range, even if we aren't quite sieved far enough. P=15T would be 3 times further sieved than the same n-range for the 1st drive so that's sufficient as far as I'm concerned.[/quote]
Hmm...why bother starting a whole new 10th Drive for the 500K-1M range? As long as we'll still be doing the entire k=1400-2000 range all all at once, we may as well just extend the 8th Drive instead--it would be a lot easier. After all, the only reason why we reset the drive numbers for k=400-1001 was because we were splitting it up into smaller k-ranges for n>600K. :smile:

gd_barnes 2009-02-25 05:48

[quote=mdettweiler;163928]Hmm...why bother starting a whole new 10th Drive for the 500K-1M range? As long as we'll still be doing the entire k=1400-2000 range all all at once, we may as well just extend the 8th Drive instead--it would be a lot easier. After all, the only reason why we reset the drive numbers for k=400-1001 was because we were splitting it up into smaller k-ranges for n>600K. :smile:[/quote]


Actually, no, I had planned on starting a new drive anyway for k=400-1001, regardless of splitting the k-ranges, because n=320K-1M would have been a huge range for 300 k's.

Like with the 1st drive, even if we don't split the k-ranges for k=1400-2000, extending it to n=350K-1M would make it too huge.

There are two things to consider here:
1. We know that Benson will take k=1003-1400 to n=600K. Will he take that range for n>600K? I think he is but I need to ask specifically.
2. Do we want to take k=1400-2000 to n=600K and then split it into 3 k-ranges or keep it as one k-range? My preference is one huge k-range. Otherwise, we would have so many drives going, it would become unmanageable.

Assuming that we keep k=1400-2000 as one huge k-range up to n=1M and if Benson will be searching k=1003-2000 for n>600K, we should create a 10th drive for n=500K-1M. Otherwise it's just too huge. If we decide to split it up into 3 k-ranges, we should extend the 8th drive to n=600K, and then create 10th/11th/12th drives for n=600K-1M.

One more caveat: If Benson is not going higher than n=600K, we should probably extend the 8th drive to n=600K, create a 10th drive for k=1003-1400/n=600K-1M and an 11th drive for k=1400-2000/n=600K-1M. I'd like to have them be different sizes like that so that they could progress at different rates, giving larger choices of n-ranges to search for people. Having one drive for 500 k's for n=600K-1M would just be too big. We like to feel like we are completing large chunks of work.

I think Karsten would scream if we did 3 more drives just for k=1400-2000, regardless of what Benson is doing with k=1003-1400 for n>600K. Also, I don't want to mess with so many different servers.


Gary

mdettweiler 2009-02-25 07:06

[quote=gd_barnes;163933]Actually, no, I had planned on starting a new drive anyway for k=400-1001, regardless of splitting the k-ranges, because n=320K-1M would have been a huge range for 300 k's.

Like with the 1st drive, even if we don't split the k-ranges for k=1400-2000, extending it to n=350K-1M would make it too huge.

There are two things to consider here:
1. We know that Benson will take k=1003-1400 to n=600K. Will he take that range for n>600K? I think he is but I need to ask specifically.
2. Do we want to take k=1400-2000 to n=600K and then split it into 3 k-ranges or keep it as one k-range? My preference is one huge k-range. Otherwise, we would have so many drives going, it would become unmanageable.

Assuming that we keep k=1400-2000 as one huge k-range up to n=1M and if Benson will be searching k=1003-2000 for n>600K, we should create a 10th drive for n=500K-1M. Otherwise it's just too huge. If we decide to split it up into 3 k-ranges, we should extend the 8th drive to n=600K, and then create 10th/11th/12th drives for n=600K-1M.

One more caveat: If Benson is not going higher than n=600K, we should probably extend the 8th drive to n=600K, create a 10th drive for k=1003-1400/n=600K-1M and an 11th drive for k=1400-2000/n=600K-1M. I'd like to have them be different sizes like that so that they could progress at different rates, giving larger choices of n-ranges to search for people. Having one drive for 500 k's for n=600K-1M would just be too big. We like to feel like we are completing large chunks of work.

I think Karsten would scream if we did 3 more drives just for k=1400-2000, regardless of what Benson is doing with k=1003-1400 for n>600K. Also, I don't want to mess with so many different servers.


Gary[/quote]
Ah, okay, that makes sense now. Thanks for explaining. :smile:

IronBits 2009-02-25 09:17

When are we recycling the drive numbers so they start over at drive 0 ? :wink:
Do we really have 10 drives going at once now?

gd_barnes 2009-02-25 09:41

[quote=IronBits;163942]When are we recycling the drive numbers so they start over at drive 0 ? :wink:
Do we really have 10 drives going at once now?[/quote]

Ha ha. Of course the drives will keep accumulating. There wouldn't be a reason to start over. For historical reference, it would create confusion.

No, drives 1 thru 4 are complete. Drives 5 thru 9 are ongoing so 5 "formal" drives that are run on servers. There is also the double-check drive and the individual-k drive, neither of which are run on servers. The latter of those 2 drives also contains a subset called the "6k mini drive" that is an informal drive on just 6 k's running on my port 8000. There is also this sieving drive but that's not really a testing drive.

Since the mini drive is a subset of the individual-k drive, we have 7 full drives currently running (5 full drives on servers) plus this sieving drive.

Drive 8 will finish and roll right into drive 10. It will be the same as when drive 1 finished and it rolled right into drives 5/6/7.

This is the reason that I don't want any one drive to be too huge. I don't want to get 10 "formal" drives going at once. We'll feel like we're never completing anything. We also may get these temporary "fill in" thingies going at times like we've had on ports 2000 and 3000 but those should generally be small.


Gary

gd_barnes 2009-02-26 21:19

My sieving range of P=1T-4T is complete. Unfortunately I don't have time right now to combine all the factors, remove them from the sieve file, and start another sieving range right now so I'm running my 3 quads on port G8000 for a day or so until I have time to get it taken care of.

gd_barnes 2009-02-27 12:56

A new file will all factors removed up to P=4000G (4T) has been Emailed to everyone.

Reserving P=6250G-9250G. My machines will start on it later today.

PCZ 2009-03-03 09:30

Ranges i have already reserved will be finished today.
I'll get them emailed to Gary tonight.

I will then sieve 9250G - 10T. Using the new file sieved to 4T

Flatlander 2009-03-03 13:26

[quote=Flatlander;163837]ETA 3rd March.[/quote]
Now 4th March due to power cut.

gd_barnes 2009-03-04 11:25

[quote=PCZ;164480]Ranges i have already reserved will be finished today.
I'll get them emailed to Gary tonight.

I will then sieve 9250G - 10T. Using the new file sieved to 4T[/quote]


Per Email received from Brian, his ranges are now complete.

Brian, can you give us an ETA on your new range?


Gary

Flatlander 2009-03-04 14:47

1 Attachment(s)
6000G-6250G complete.
14182 factors found.
Taking 10000G-10250G. (Using the new file sieved to 4T.)


[B]edit[/B]: ETA for 10000G-10250G, 19th March.

MyDogBuster 2009-03-04 18:53

I've freed up 3 FAST cores.

Taking 10250G to 11000G. ETA 21 Mar

Note: The slow ones are still Mar 19th

PCZ 2009-03-04 21:55

9250G to 10T ETA 8th Mar

PCZ 2009-03-08 22:14

9250G to 10T Completed emailed to Gary.

Sieving 11T - 12.5T
ETA 16th Mar

gd_barnes 2009-03-15 10:52

P=6250G-9250G is complete. Factors sent to myself.

Reserving P=12500G-15000G. ETA is March 30th.

I'll continue sievng P=250G per core so will pull a couple of cores off to put on port 5000 for a while.

At P=15000G (15T), that will be a sufficient depth for n=500K-600K, so we can start a new drive for k=1400-2000 for n=500K-1M with just that range to start with (after the 8th drive is done). We could also run a rally or two on the new drive to spike a large # of top-5000 primes and possibly attract some more new folks to our effort. :smile: (Note: Peter Benson has indicated that he will continue with k=1000-1400 past n=600K.)

Once I have all of the factors up to P=10T, I'll remove them from the file, Email everyone the updated file, and compute an optimum sieve depth for the entire range. We'll leave n=500K-600K in there since continuing to sieve it will save time on future double-checking.


Gary

PCZ 2009-03-17 18:15

Revised ETA for 11 - 12.5 T
18th March.

Flatlander 2009-03-19 20:23

1 Attachment(s)
10000-10250G complete.
8131 factors found. Expected c. 8017.25

MyDogBuster 2009-03-19 23:32

4000G-4250G
4500G-4750G
5000G-5250G
5500G-5750G

All complete results emailed No primes found

NOTE: That machine heading for the repair bench

MyDogBuster 2009-03-20 18:19

Reserving 15000G - 16000G

Also 10250G - 11000G Complete Results emailed

gd_barnes 2009-03-21 08:17

[quote=MyDogBuster;166037]4000G-4250G
4500G-4750G
5000G-5250G
5500G-5750G

All complete results emailed No primes found

NOTE: That machine heading for the repair bench[/quote]


Well, if you sieved up to P=~sqrt(1401*2^500001-1), you might have found some primes. I think you need to get busy.

gd_barnes 2009-03-25 01:21

On March 18th, Bryan reported completion of the P=11000G-12500G range and sent me the factors.

I'll be sending everyone a new sieve file here in a little while with all factors removed up to P=12500G (12.5T).


Gary

gd_barnes 2009-03-25 02:46

I have Emailed a new file sieved to P=12.5T to Ian, Bryan, and Chris. If anyone else would like to help sieve, let me know and I'll send you the file.


Gary

PCZ 2009-03-25 07:37

I will sieve 16T -17T
ETA 4th April

Flatlander 2009-03-25 12:52

Taking 17-17.25T.
ETA 9th April

Brucifer 2009-03-27 15:44

I'll go ahead and take it from 17.25T on up to finish per chat with Gary.

AMDave 2009-03-27 16:07

[QUOTE=MyDogBuster;166037]NOTE: That machine heading for the repair bench[/QUOTE]
It was worth it!

MyDogBuster 2009-03-27 16:36

[QUOTE]
I'll go ahead and take it from 17.25T on up to finish per chat with Gary.
[/QUOTE]

I'm going to ASS U ME, you are taking it to 30000G. If so, nice going Mr Brucifer. I've heard rumors of your legendary prowess as a siever.

Flatlander 2009-03-27 17:48

[quote=Brucifer;166870]I'll go ahead and take it from 17.25T on up to finish per chat with Gary.[/quote]
We are very appreciative. :bow:

Brucifer 2009-03-27 18:24

Just waiting for Gary to send me the current dat file and then I'll load some systems up.

Brucifer 2009-03-27 18:26

[QUOTE=MyDogBuster;166879]I'm going to ASS U ME, you are taking it to 30000G. If so, nice going Mr Brucifer. I've heard rumors of your legendary prowess as a siever.[/QUOTE]

U R ass u ming correctly. As for legendary..... LOL maybe a legend in my own mind.... ROFLMAO. :smile:

gd_barnes 2009-03-28 17:38

[quote=Brucifer;166896]Just waiting for Gary to send me the current dat file and then I'll load some systems up.[/quote]

The file has been sent. I could not easily send a .tar version from my Windows laptop while I'm out of town with my Windows laptop so I sent a .zip version. If a .tar version is needed, just post that here and I'm sure Ian, Chris, or Bryan could send you one. As I recall, I could pull up a .zip version in Linux so hopefully that will be OK.

My 2.5T range will be done here in 3 days. I'll be glad to move the 10 cores (originally 12) back to primality testing.

Thanks a bunch Bruce! :smile:


Gary

Brucifer 2009-03-28 23:50

@gary -- yup the file worked. The sieving is moving forward.. Don't thank me until it's finished. :smile:

gd_barnes 2009-03-30 17:56

Bruce estimated 9 weeks to complete the sieving from his Mar. 29th Email. I'll add a few days to allow for any issues and show an ETA of June 3rd. That should work well.

gd_barnes 2009-04-01 07:10

I've changed the completion date on my range to April 5th due to my power outage.

Also, I've Emailed everyone a smaller sieve file. When removing factors up to P=12.5T, I forgot to remove some smaller factors.

Sorry about the oversight.


Gary

gd_barnes 2009-04-04 17:34

FINALLY last night P=12.5T-15T completed. Factors sent to myself.

Max, Karsten, and Ian, since P=15T-17T should be done in just a few days, I'll wait and remove the P=12.5T-17T factors from the "big" file. We'll then use a file sieved to P=17T for n=510K-600K unless port 8000 looks like it might dry before those are done. In that case, we'll use the P=12.5T file for n=500K-520K and P=17T file for n=520K-600K. (Note: I'm only skipping the P=17T-17.25T file that may also be done to keep things more round.)

With the way we're going on port 8000, if it looks like we'll hit n=600K on the drive before we finish sieving to P=30T, like before, I'll pull all of my machines off of it and either assist with sieving or work on other ports.

The plan at this point is to have 3 sieve depths and no more for the 10th drive. It's amazing how much faster we have progressed on it then the 1st drive for k=400-1001 and n=260K-600K, which had nearly the same # of k's. The big BBQ rally made a big difference there.


Gary

Brucifer 2009-04-04 19:17

Gary,
By tonight I will have loaded into the systems up thru 19750e9. I mention that in case you care to go back to using the reservation method here in case you are going to go back and help on sieving. That way we wouldn't get our wires crossed, and if anyone else wanted to take some ranges they could do so also. That should get the sieving done much faster then.

Around 8pm PST time today I will send you the factors for 17750e9 - 18750e9. I previously sent you the factors for 17250e9 -17750e9 and am assuming you received those okay.

Bruce

gd_barnes 2009-04-06 01:43

[quote=Brucifer;168061]Gary,
By tonight I will have loaded into the systems up thru 19750e9. I mention that in case you care to go back to using the reservation method here in case you are going to go back and help on sieving. That way we wouldn't get our wires crossed, and if anyone else wanted to take some ranges they could do so also. That should get the sieving done much faster then.

Around 8pm PST time today I will send you the factors for 17750e9 - 18750e9. I previously sent you the factors for 17250e9 -17750e9 and am assuming you received those okay.

Bruce[/quote]


Yes, I received both the 17250e9-17750e9 and 17750e9-18750e9 groups of factors. Thanks.

If you don't mind continuing to sieve, I'll just keep you reserved for the entire range. If it gets old, please say so and I'll open back up the sieving reservations.

We're pumping out the top-5000 primes on the 10th drive now so I'm quite excited about keeping that going almost full-bore at least through the rally.


Gary

Brucifer 2009-04-06 04:14

The sieving is a good thing for me at present as I seem to have developed a heck of an intermittent connection issue with the ISP that has been costing me a few hours of no connections. So it looks like the games will begin with the trouble desk and the incompetent ones. :(

MyDogBuster 2009-04-06 05:32

15000G - 16000G complete

Results emailed

PCZ 2009-04-06 07:55

16T-17T is going to be finished in about 36hrs

Brucifer 2009-04-06 16:49

Gary --when you have all the sieving ranges from others completed, could you send me a new updated dat file please? :smile:

gd_barnes 2009-04-06 18:43

[quote=Brucifer;168293]Gary --when you have all the sieving ranges from others completed, could you send me a new updated dat file please? :smile:[/quote]

Yes, that was the intent. It looks like they should be done late Tuesday so I'll send it over shortly after I get them.

Flatlander 2009-04-06 18:49

[quote=Flatlander;166634]Taking 17-17.25T.
ETA [B]9th April[/B][/quote]
:smile:

PCZ 2009-04-07 21:04

16T-17T completed emailed to Gary.

gd_barnes 2009-04-07 22:52

[quote=PCZ;168420]16T-17T completed emailed to Gary.[/quote]

Brian,

I just sent you an Email. There are gaps in your factors at the end of each P=250G range. It looks like your 4 cores weren't quite done yet.

I'll mark the range as not yet complete with a completion date in a couple of days.


Thanks,
Gary

PCZ 2009-04-08 06:22

Gary sorry about that.
Sent you partially completed work by mistake.
I have sent another mail to you with the completed ranges.

gd_barnes 2009-04-08 07:42

[quote=PCZ;168458]Gary sorry about that.
Sent you partially completed work by mistake.
I have sent another mail to you with the completed ranges.[/quote]


Great. Thanks. It looks good. I'll note the range complete.

gd_barnes 2009-04-08 23:26

Reserving P=19.75T-19.85T for work on one quad. ETA is Apr. 10th PM EDT.

Bruce and I are coordinating on this. We're going to try to get the file sieved to P=20T at least a 1/2-day before the next range completes in port 8000.

David, we'll be cutting it a little close. I sent you an Email with the particulars.

Bruce, I have you estimated to complete P=18750e9-19750e9 by early morning (EDT) on Apr. 9th per a previous Email from you. I then put you down for early morning (EDT) on Apr. 11th for completing 19850e9-20000e9.

Chris, just to confirm...you'll be finishing your range sometime on Apr. 9th, correct?

We'll then use the P=20T sieved file for n=520K-600K for the 10th drive.


Gary

Brucifer 2009-04-09 00:03

[QUOTE=gd_barnes;168550]
Bruce, I have you estimated to complete P=18750e9-19750e9 by early morning (EDT) on Apr. 9th per a previous Email from you. I then put you down for early morning (EDT) on Apr. 11th for completing 19850e9-20000e9.

Gary[/QUOTE]

I shot you an email on this last night and then one today regarding stuff to 20T. Didn't hear back. I've got 19750e9-20000e9 loaded in to one core, and the rest on that machine also have 250G ranges loaded in. They complete April 29th.

If you want the stuff to 20T then I sure wish you would have said something a little sooner. Last you told me was you were just going to use up to 18T and load that in to the server and not worry about the rest. grrrrrrrrrrrrr :)

I can load the stuff in to multiple systems to get the 19850-20000 done faster if that's what you want. I hope you aren't going to want anything from above 20T done before the end of the month cause they are all loaded in large chunks like you mentioned in a message. Otherwise then I need to go back and modify the loads if anything above 20T is going to be needed sooner.

Flatlander 2009-04-09 00:04

[quote=gd_barnes;168550]...
Chris, just to confirm...you'll be finishing your range sometime on Apr. 9th, correct?
...
Gary[/quote]
Yes, late afternoon GMT. (Unless I get another power cut...)

Brucifer 2009-04-09 00:26

Gary I just answered your email. I split that range over two Q9400's, 8 cores. Around 9:30am PST on the 10th is when my part of the 20T will be finished except for the range you loaded on to your systems. I hope that works for you. See the email please... the one labeled the crises. :)

gd_barnes 2009-04-09 00:33

[quote=Brucifer;168558]Gary I just answered your email. I split that range over two Q9400's, 8 cores. Around 9:30am PST on the 10th is when my part of the 20T will be finished except for the range you loaded on to your systems. I hope that works for you. See the email please... the one labeled the crises. :)[/quote]

Excellent! I hadn't realized you could start on it right away and had assumed that you would have to do 25G per core instead of the 20G that you stated in your Email. I'll change your ETA to 4/10 in the AM for 19850e9-2000e9.

David, it's not so bad now. The sieving should be done to P=20T by 6-8 PM EDT on Apr. 10th. Port 8000 wouldn't dry, at its current rate, until close to noon on Apr. 11th so we're in decent shape. I'll keep an eye on its testing rate. If someone adds a bunch of cores, I'll pull some of mine over to port 5000 for a while.


Gary

Brucifer 2009-04-09 00:44

Dumb question time...... Looking down the road on the rally, is any of this stuff going to be used? If so, how much more needs to be completed. Just asking before I reload stuff after this 20T stuff is finished.

gd_barnes 2009-04-09 01:20

[quote=Brucifer;168560]Dumb question time...... Looking down the road on the rally, is any of this stuff going to be used? If so, how much more needs to be completed. Just asking before I reload stuff after this 20T stuff is finished.[/quote]

I'm a little confused by your question but will answer what I think you're after:

Once we hit P=20T (20000e9), we'll use those factors for the n=520K-600K range files of the 10th drive and so it will be quite a while before we need any more factors. The rally will not come even close to getting up to n=600K even if we have the same # of resources as the last huge rally. (Not likely because it appears that Beyond is no longer available.) I'm hoping we'll near n=550K by the end of the rally, which would be a substantial accomplishment over the next 10 days or so, but n=540K would be very good also.

Therefore, you can take your time with 20T-30T because we won't use those factors until the 10th drive is nearing n=600K. I would suggest loading something like P=250G ranges into each core so you aren't messing with them stopping and having to restart new ranges frequently. Also, it's easier for me to handle fewer large factor files than many smaller files.

We're leaving the entire n-range in the file and continuing to sieve it to P=30T because it takes little additional time to sieve n=500K-1M vs. 600K-1M. We will use the deeper sieved file for double-checking in the future, which will save time there.

To go math-geek on you: Sieving n=500K-1M vs. 600K-1M takes sqrt(500K/400K) longer = 1.118 = 11.8% more time and will save us more than that on double-checking in the future.

Future double-checking is also why we have the k=1003-1400 range in there even though we're not currently testing it. But there's another reason: It's possible that Benson may not want to take k=1003-1400 all the way up to n=1M (he's stated in an Email that he is taking it to n>600K but hasn't specified how high yet) so we will be here to pick up where he left off if he stops.

In other words, we're trying to cover all of the possibilities in the future by continuing to sieve the entire k and n-range.

Does that answer what you are after or way more than you're after? :smile:


Gary

gd_barnes 2009-04-09 02:52

Ah, we're in even better shape than I thought. My prior estimate was extrapolated from when I was sieving far lower P-ranges. With far less pairs in the file now, the sieving is faster. Right from the source itself, sr2sieve, my sieving range should complete about 7 AM Friday morning CDT.

David, we'll have more than a day leeway on the next file for port 8000 now.

Brucifer 2009-04-09 06:46

[QUOTE=gd_barnes;168564]
Does that answer what you are after or way more than you're after? :smile:

Gary[/QUOTE]

yup. :)

Flatlander 2009-04-09 14:20

1 Attachment(s)
17-17.25T complete.
[quote]Found factors for 5355 terms in 1294896.026 sec. (expected about 5358.63)[/quote]

Brucifer 2009-04-10 01:05

Yo Gary... the faster one is now averaging 224000 p/sec per core. The slowest is doing 203/204 so big change. I wasn't using v1.7.15, so updated them. Changes the projections by days like from a completion on 1 May down to 22 April for a 250G chunk per core. So they are running like yours now. :)

gd_barnes 2009-04-10 04:41

[quote=Brucifer;168696]Yo Gary... the faster one is now averaging 224000 p/sec per core. The slowest is doing 203/204 so big change. I wasn't using v1.7.15, so updated them. Changes the projections by days like from a completion on 1 May down to 22 April for a 250G chunk per core. So they are running like yours now. :)[/quote]


Excellent! I'm glad we checked your P-rate. Once I get all the factors up to P=20T on Friday, I'll compute the optimum sieve depth and send you a new file.

For everyone's reference: It's looking like we'll probably need to sieve this range to P=40T-50T somewhere. A bad error on the original estimate on my part. With both the k-range and n-range being larger than any previous sieving effort, there is quite a bit to be gained in total testing time to sieve deeper than other efforts. Based on that, I'll probably open back up the sieving drive here. As for me, after the end of the rally, I'll probably keep one full quad on it full time until it's done.

One final note: Bruce reported completion of P=18750G-19750G in an Email.


Gary

IronBits 2009-04-10 04:58

If I were to take P=40-50T range...
If you break it up to go across 4 x 8 cores = 32 * 3.5 GHz cores
How long would that take?

gd_barnes 2009-04-10 05:18

[quote=IronBits;168708]If I were to take P=40-50T range...
If you break it up to go across 4 x 8 cores = 32 * 3.5 GHz cores
How long would that take?[/quote]

I'm not sure yet that we will need to go all the way to P=50T. What I meant was that we may need to sieve up to BETWEEN P=40T and 50T somewhere.

But...if you would like to take the range of P=30T-40T, that would be excellent! We'll almost definitely need that.

Let's see: P=250G (0.25T) on one 2.6 Ghz core takes right around 15 days right now so on 32 3.5 Ghz cores, you could do:

32 * 0.25T * 3.5 / 2.6 = 10.77T in 15 days.

Well, isn't that convienient that it comes right out close to a P=10T range? :smile: Therefore assuming that you have a 64-bit machine, O.S., and the 64-bit version of sr2sieve (which I'm sure you do, lol), I'm confident that you could complete P=30-40T in 15 days on 32 cores.

If you'd like that range, I'll reserve it for you and then send you the new sieve file late tomorrow afternoon after we have all of the factors up to P=20T removed (at the same time that I send you another file for port 8000). You could feel free to take a break for the rally. :smile:

If Bruce does P=20T-30T, you do P=30T-40T, and the optimum sieve depth turned out to be P=50T, I'm sure that between you guys, me and the rest of the NPLB group, we could all get it done by early June, which is what I'm hoping for. (I think the 10th drive will be nearing n=600K at that point.)


Gary

IronBits 2009-04-10 05:43

After the rally, sign me up for 30-40T then.

If you or Max could divide it up for 4 computers with 8 cores each, then I'll fire em up.
(like last time) :wink: except this time, I'll let it finish correctly :grin:

mdettweiler 2009-04-10 06:23

[QUOTE=IronBits;168712]After the rally, sign me up for 30-40T then.

If you or Max could divide it up for 4 computers with 8 cores each, then I'll fire em up.
(like last time) :wink: except this time, I'll let it finish correctly :grin:[/QUOTE]
Okay, I'll see about setting you up with a "care package" like last time as soon as the p=20T sieve file becomes available. Remind me when the time comes if I forget... :smile:

gd_barnes 2009-04-10 06:53

[quote=IronBits;168712]After the rally, sign me up for 30-40T then.

If you or Max could divide it up for 4 computers with 8 cores each, then I'll fire em up.
(like last time) :wink: except this time, I'll let it finish correctly :grin:[/quote]

Excellent! Thanks for helping out.

Bruce, although I put you down for P=20T-30T, after you get your machines rolling with the next P>20T range (I'm assuming 250e9 per core), let me know what specific range you are running since I'm not clear exactly how many cores you will be sieving with. I will likely put the end of your range up to P=30T up for public reservation. Therefore, ultimately one of the ranges running on your machines at one time may end up having a big gap in the middle, perhaps like P=28T-30T and 40T-42T.

Max, I'll send the P=20T sieve file to you and you can include it in your care package to David. As for dividing P=10T by 32 cores, it would be 0.3125T or 312.5G or 312500M per core. Since that's a bit messy, you might make half of them 312G and the other half 313G. If you do that, I would suggest making the higher ranges 313G since they sieve a tad faster. So the 1st core would be 30000e9-30312e9, 2nd core, 30312e9-30624e9, etc. That should put them done all at very close to the same time.


Gary

mdettweiler 2009-04-10 06:59

[QUOTE=gd_barnes;168715]Max, I'll send the P=20T sieve file to you and you can include it in your care package to David. As for dividing P=10T by 32 cores, it would be 0.3125T or 312.5G or 312500M per core. Since that's a bit messy, you might make half of them 312G and the other half 313G. If you do that, I would suggest making the higher ranges 313G since they sieve a tad faster. So the 1st core would be 30000e9-30312e9, 2nd core, 30312e9-30624e9, etc. That should put them done all at very close to the same time.[/QUOTE]
Okay, that sound good. Having the ranges slightly differing in size as you described should definitely make things a little easier to keep track of, and any resulting differences in runtime should surely be within a normal range of variation that occurs due to other processes on the system and other such factors. :smile:

Brucifer 2009-04-10 15:47

Mr. Barnes, you will find in your email inbox 19850e9-20000e9 in gzip format.

I have now loaded in 20000e9 to 25000e9 in 250G ranges, covering 20 cores. These should complete on Apr 23/24.

So let me know what your plan is for the remaining ranges when you decide please. :smile:

gd_barnes 2009-04-10 20:58

P=19750G-19850G is complete.

We are now complete up to P=20T. A new sieve file has been sent to everyone.

Bruce, since it appears that you are running 20 cores for sieving, would you want to just put a P=500G range in each of them? That would take care of the P=20T-30T range and would make things basically "maintenance free" on your end. That would take you just a little under 30 days. I've noted the range as reserved by you with an ETA of May 15th, allowing 5 days for unforseen problems and running the rally.

I'm now preparing the next file for port 8000 and will run some tests to verify the optimum sieve depth.


Gary

gd_barnes 2009-04-10 22:32

Well, bad news followed by good news...

Bad news:

The optimum sieve depth for the entire range is definitely right at P=60T. (ouch) This is based on a current factor remove rate of 310 secs. and a test at 70% of the n-range (n=850K) taking 920 secs.

Good news:

We're not testing k=1003-1400 at the current time and the continued sieving of the n=500K-600K range is only for future double-checking.

Although we will NOT be removing the above portions of the sieve file, because the double-checking of both of those ranges will be well into the future when computer capacity is much greater, I feel that it should be taken into account.

Therefore we have:

3/5ths of the k-range * 4/5ths of the n-range = 12/25ths of the entire file will be tested in the foreseeable future.

The square root of 12/25 = ~.7

60T * .7 = 42T

So let's just round it up and sieve to P=45T. If we double-check in the future, if we want, we can sieve it further because capacity and speed will be much greater.

Sieving with today's resources to P=60T is overkill. Even if we were testing the entire range, it will be so long before it is complete that we could always sieve further in the future for the higher n-ranges.


Gary

IronBits 2009-04-10 23:18

Sign me up for 30-45T then.
What's an extra day or two? :wink:


All times are UTC. The time now is 14:49.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.