Mersenne Digest Saturday, February 7 2004 Volume 01 : Number 1105
---------------------------------------------------------------------- Date: Sat, 17 Jan 2004 10:21:13 -0500 (EST) From: [EMAIL PROTECTED] Subject: RE: Mersenne: p95 dear friends , I want to know if it is possible to distribute LL algorithm so that we can run it on a cluster to test one exopnent on several nodes simultaneously so check it in a shorter peroid of time . Or should we use trial_factoring algorithm in this regard!. or other algorithms.??? > [EMAIL PROTECTED] >> to whom it may concern : >> my question is: will the p95 software run on a system of >> clusters?????? > > If you're asking if there's a cluster-aware version of Prime95, the > answer is no. Because of the nautre of the error checking done on the > server, there is no need to provide failover of the service. Nor is > there a need to have two coordinated processes running on each node to > increase performance - each node can run a copy of P95 independently, > testing different exponents while maintaining the maximum performance > possible. > > You can use prime95 on a cluster by running a separate copy on each > node, and you'll get all the performance your hardware can provide. If > your particular cluster architecture requires "mirror image" execution > and file-systems, then you'll have a problem, because both nodes will > perform exactly the same prime95 computations, and you'll get the same > performance as a single machine. > > Regards, > Ryan > _________________________________________________________________________ > Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm > Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers > _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Fri, 16 Jan 2004 23:40:38 -0800 From: "John R Pierce" <[EMAIL PROTECTED]> Subject: Re: Mersenne: p95 > I want to know if it is possible to distribute LL algorithm so that we can > run it on a cluster to test one exopnent on several nodes simultaneously > so check it in a shorter peroid of time . Or should we use trial_factoring > algorithm in this regard!. or other algorithms.??? not doable. there's just WAY too much data exchanged in each pass. each pass of the FFT used in the Lucas Lehmer algorithm crunches through something like 8MB of transform data in around 0.05 seconds (on my p4-2.5Ghz). 21 million passes later, its done. This takes about 10-12 days. If you have a cluster of 100 such processors, the best possible thing they can do is each work on a different exponent for 10-12 days, then you'll get 100 times the throughput of a single processor. _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sat, 17 Jan 2004 07:44:46 +0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: double-check mismatches On Saturday 17 January 2004 02:32, Daran wrote: > On Thu, Jan 15, 2004 at 07:15:46PM +0000, Brian J. Beesley wrote: > > ...matching > > residuals mean that the chance of an error getting into the database as a > > result of a computational error is of the order of 1 in 10^20. > > That's per exponent, isn't it? The chance that one of the roughly quarter > million status-doublechecked exponents being in error is about five orders > of magnitudes higher. Sure. That's why I ran the project to triple-check a not inconsiderable number of smaller exponents where one (in some cases both) of the residues was reported to less than 64 bits, usually only 16. No discrepancies were discovered. > > Still acceptible, or at least a minor consern in comparison to the other > security issues. > It's easy enough - and computationally exceedingly cheap - to report more residue bits but, as you say, other issues are not so easy to fix. Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sat, 17 Jan 2004 07:44:46 +0000 From: "Brian J. Beesley" <[EMAIL PROTECTED]> Subject: Re: Mersenne: double-check mismatches On Saturday 17 January 2004 02:32, Daran wrote: > On Thu, Jan 15, 2004 at 07:15:46PM +0000, Brian J. Beesley wrote: > > ...matching > > residuals mean that the chance of an error getting into the database as a > > result of a computational error is of the order of 1 in 10^20. > > That's per exponent, isn't it? The chance that one of the roughly quarter > million status-doublechecked exponents being in error is about five orders > of magnitudes higher. Sure. That's why I ran the project to triple-check a not inconsiderable number of smaller exponents where one (in some cases both) of the residues was reported to less than 64 bits, usually only 16. No discrepancies were discovered. > > Still acceptible, or at least a minor consern in comparison to the other > security issues. > It's easy enough - and computationally exceedingly cheap - to report more residue bits but, as you say, other issues are not so easy to fix. Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Sat, 7 Feb 2004 00:30:22 -0800 (PST) From: Emrecan Dogan <[EMAIL PROTECTED]> Subject: Mersenne: Factoring above 2^67 - --0-1418547532-1076142622=:5853 Content-Type: text/plain; charset=us-ascii Hi, One of my computers is assigned to run factoring tests continuously. I saw that my that computer's last two results were "no factor up to 2^67" . Then i checked the "cleared exponents list" on the primenet webpage. There were no trace of my exponents but i saw many factored exponents with a factor bound to 2^101, 2^71, 2^68, 2^86 and so on. What is the criteria that an exponent will be tested up to which power? What must i do if i want to continue testing my exponent above 2^67. Bye - --------------------------------- Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online - --0-1418547532-1076142622=:5853 Content-Type: text/html; charset=us-ascii <DIV>Hi,</DIV> <DIV> </DIV> <DIV>One of my computers is assigned to run factoring tests continuously. I saw that my that computer's last two results were "no factor up to 2^67" . Then i checked the "cleared exponents list" on the primenet webpage. There were no trace of my exponents but i saw many factored exponents with a factor bound to 2^101, 2^71, 2^68, 2^86 and so on.</DIV> <DIV> </DIV> <DIV>What is the criteria that an exponent will be tested up to which power? What must i do if i want to continue testing my exponent above 2^67. </DIV> <DIV> </DIV> <DIV>Bye</DIV><p><hr SIZE=1> Do you Yahoo!?<br> Yahoo! Finance: <a href="http://us.rd.yahoo.com/evt=22055/*http://taxes.yahoo.com/filing.html">Get your refund fast by filing online</a> - --0-1418547532-1076142622=:5853-- _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ End of Mersenne Digest V1 #1105 *******************************