Mersenne Digest Wednesday, October 31 2001 Volume 01 : Number 898 ---------------------------------------------------------------------- Date: Mon, 29 Oct 2001 19:30:54 +0100 From: Henk Stokhorst Subject: Re: Mersenne: number of processors participating Alan Vidmar wrote: > I suggest that there be a switch added so >that ppl can use Prime95 as a processor test but without ever >getting real assignments,... > This is a VERY good suggestion. However it has already been implemented in the latest version (v21). That version contains more improvements so I wondered if it wouldn't be a good idea to inform users through the occasional newsletter. Particulary because it gives a 10% improvement for Pentium I, II and III users and it skips P-1 if it has been done. YotN, Henk Stokhorst. PS those abandoned assignments do't slow down the project. They just scatter the work over a larger range. _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 22:01:48 +0100 From: =?utf-8?Q?Torben_Schl=C3=BCntz?= Subject: SV: Mersenne: number of processors participating I admit I'm not that good in telling primenet what computers I have and what throughput rate to expect. eg.: I made 14 accounts all using the same 150 Mhz machine, though I knew none or only few would be 150 Mhz. These accounts all run occassionally, eg. in company holiday around the clock, outside of holiday more random. Over time I have been wiser to use more power of those machines staying awake all night anyway. :-) I would like to use the servers; but I haven't been able to persuade George to make a Quit function like quit_at: 06:00 to terminate the program when users arrives and optimum performance is needed (with no question what so ever about serverperformance); And I don't wake up at 6 to turn prime95 or anything else off unless there is a severe error reported by users. Happy hunting tsc -----Oprindelig meddelelse----- Fra: Henk Stokhorst Sendt: ma 29-10-2001 19:30 Til: Alan Vidmar; mersenne@base.com Cc: Emne: Re: Mersenne: number of processors participating Alan Vidmar wrote: > I suggest that there be a switch added so >that ppl can use Prime95 as a processor test but without ever >getting real assignments,... > This is a VERY good suggestion. However it has already been implemented in the latest version (v21). That version contains more improvements so I wondered if it wouldn't be a good idea to inform users through the occasional newsletter. Particulary because it gives a 10% improvement for Pentium I, II and III users and it skips P-1 if it has been done. YotN, Henk Stokhorst. PS those abandoned assignments do't slow down the project. They just scatter the work over a larger range. ________________________________________________________________________ _ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 22:59:18 +0000 (GMT) From: Russel Brooks Subject: Re: Mersenne: Statistics Achim.Passauer@RegTP.DE wrote: > I mentioned that we lost about 9000 machines during the past 7 months. One > year ago I started collecting data about numbers of machines, accounts and > so on. Please find this information at the bottom of may mail. GIMPS has now > as many contributing machines as it had 13 months ago. In the meantime there > was a peak of 38950 machines (26 March 2001). After that date we lost about > 9000 machines which we had gained between September 2000 and March 2001. As Since March 2001? Isn't that about when California's power crisis started warning about summer blackouts? Maybe a lot of the missing machines were CA users shutting down their pcs. I left mine on. Cheers... Russ (in San Jose) _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 00:23:20 +0100 From: =?utf-8?Q?Torben_Schl=C3=BCntz?= Subject: SV: SV: Mersenne: number of processors participating Yep! But the time entry only allows the program to sleep (still eating all CPU cycles even when running at zero priority). Take any NT 4.0 or W2K machine and you will see the "system idle time" doesn't add seconds while Prime95 still eats them (and doing nothing). For my servers to become prime95's I need to be sure they only run what I have planned at anytime. I can start Prime95 scheduled. I don't mind! But the users should never have one chance of claiming servers aren't available or even running "slow". I know you are certain and I know you gotta be damn good at this (very far beyond anything I will ever manage); but still any doubt will become my users advantage. Make the sleepy nights for my servers glorius. I make them start prime95 by a schedule and You make prime95 die by harikiri - and I decide when everything happens. :-) Tnx in advance. Still happy hunting tsc -----Oprindelig meddelelse----- Fra: George Woltman Sendt: ma 29-10-2001 22:47 Til: Torben Schlüntz Cc: Emne: Re: SV: Mersenne: number of processors participating Hi, At 10:01 PM 10/29/2001 +0100, you wrote: >I would like to use the servers; but I haven't been able to persuade >George to make a Quit function like > > quit_at: 06:00 > >to terminate the program when users arrives and optimum performance is >needed Look in readme.txt for the Time= entry in prime.ini This feature can be used to make prime95 go dormant at a specified time. Hope that helps, George _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 23:46:45 -0000 From: bjb@bbhvig.uklinux.net Subject: Re: Mersenne: number of processors participating On 29 Oct 2001, at 19:30, Henk Stokhorst wrote: > [... snip ...] However it has already been > implemented in the latest version (v21). That version contains more > improvements so I wondered if it wouldn't be a good idea to inform > users through the occasional newsletter. Particulary because it gives > a 10% improvement for Pentium I, II and III users and it skips P-1 if > it has been done. Umm - I haven't noticed any significant improvement in v21 speed on Pentium or Pentium II - the big changes are implementing prefetch, which is only applicable to AMD Athlon family, PIII and faster Celeron processors, and exploiting the SSE2 instruction set on Pentium 4 only. Apart from (sometimes) skipping P-1, the changes between v20 and v21 are pretty well cosmetic if you're using a 486 / Cyrix / AMD K6 / Intel Pentium (Classic or MMX) / Pentium Pro / Pentium II / Celeron < 533 MHz CPU. There _are_ some other changes - including a bit of fine tuning of the exponent / FFT run length size breaks - but nothing which really makes an upgrade look inescapable. In fact, these older systems are more likely to have a memory constraint than newer systems with faster processors; due to the inclusion of the Pentium 4 specific SSE2 code, the v21 binary has a significantly bigger memory footprint, so systems which won't benefit from the prefetch code & which are feeling memory pressure might be better _not_ upgrading. The speed improvement from v20 to v21 on a PIII or Athlon system should be somewhere close to 25%, rather than 10%. On these systems an upgrade seems highly desirable. If you're still running v20 (or earlier) on a Pentium 4, then quite frankly you really SHOULD upgrade. NOW. The execution speed will approximately treble. As for "reduced participation" - whilst other reasons certainly do have an effect, I've previously mentioned two other possible reasons: (1) adverse publicity stemming from the prosecution of a sysadmin for running RC5 clients on "his" systems without the agreement of the management at the college which employed him; (2) steep rises in electricity prices and unreliability of supply in some places e.g. USA West Coast deterring people from running extended jobs. Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 23:46:45 -0000 From: bjb@bbhvig.uklinux.net Subject: Re: Mersenne: number of processors participating On 28 Oct 2001, at 0:28, Terry S. Arnold wrote: > Another consideration is that many system/network administrators have > gotten ludicrous about what they will allow on their networks. They > think that Prime95 just might let in a virus or even worse spill > "company secrets". By and large they are totally ignorant of the real > issues involved with securing networks. All most of them know about > the implications of the various protocols in the TCP/IP suite was what > it took get their MCSE if they have even that much training. As a "system/network administrator" specialising in security matters I just _have_ to answer this one. 1) It's perfectly true that there are a large proportion of sites with incompetent sysadmins - especially from the point of view of networking. Especially in small companies, where the sysadmin function tends to be bolted onto another job as a low-priority "extra" task. 2) AFAIK none of the MSCE courses cover security in any depth at all. In fact the approach seems to be the _reverse_ i.e. teach people how to set up & administer systems in an unduly risky way, without even bothering to mention basic security tools or methodology because they're not essential to _Microsoft_ networking in a laboratory/classroom environment. Based on recent experiences with Code Red & Nimda, 95% of the problems on our site came from the 1% of the systems located in business incubator centres attatched to the University but "administered" by the businesses themselves. Basically it's rare for these people even to be aware of most of the services running on their systems (anything that comes preloaded on the system gets run, irrespective of whether it's absolutely neccessary or absolutely unneccessary); as for applying critical updates, they seem to be trained to think one of (a) it's much too hard, (b) it will break the functionality, (c) they simply don't understand why they need to bother with such things. _Despite_ how easy it is to run Windows Update. The only way I've been able to get these people to apply updates is to get sanctioned to scan their systems for vulnerability to Code Red & Nimda, & block _all_ access to vulnerable systems until they get patched (or take down the IIS service). To my knowledge, many ISPs had to take similar action. At least _some_ universities & Fortune 500 companies have competent sysadmins, but there are a whole lot of "mom & pop" businesses out there; a high percentage of them would be an absolute pushover to anyone "wearing a black hat", even if IIS installations have now mostly been patched to a reasonable level. As for distributed computing projects being a security risk - basically I think in many cases _management_ may be misusing "security" as a screen for filtering out anything _they_ don't understand. In my experience few of these people are aware of the scale of network _abuse_ (note, not _neccessarily_ a threat to security) that goes on by way of end users installing peer-to-peer "file sharing" software on their workstations; probably 99% of the files "shared" over these P2P networks are in effect illegal distributions of copyrighted material. They're certainly _not_ aware that Windows systems with e.g. Kazaa clients are quite capable of "sharing" not just the offending copyrighted material but also everything else on the system - or attatched to it through open LAN shares. Yes, including "company secrets". Quite apart from that, the volume of traffic involved with these P2P networks can be huge, certainly enough to seriously impact network links. (Before anyone takes me to task on the above paragraph, quite frankly I am totally opposed to the DMCA, the proposed SSSCA and all similar legislation. But I am also opposed to unauthorized distribution of copyrighted material. IMO the force of the law should be applied against those individuals making the copies, not against those who write software or the posession of hardware which might possibly be used to make illegal copies.) Under these circumstances I find it hard to understand how anyone can think that compute-intensive, network-friendly applications can be a problem. As for "letting in a virus" - if people really thought that, they just wouldn't use Microsoft products. How much of a threat was Code Red or Nimda infection on a system which wasn't running Microsoft Exchange, Microsoft Internet Information Server or (in the case of Nimda) Microsoft Internet Explorer? Well, _other_ infected systems might load up your network to some extent, but _your_ system certainly wasn't going to get infected! Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 17:05:02 -0800 From: "Aaron Blosser" Subject: Re: SV: Mersenne: number of processors participating Still the only time I've ever seen Prime95/NTPrime slow down a system is when I was doing some Netmeeting video conferences. With it running, the video conference would run DOG slow. Stop the NTPrime service and curiously had to restart the video conference for the effect, but the video would then be running great. That was with, umm.. version 20 I think? I haven't tried again with later versions... wasn't one of the things George did something to do with the priority setting? Aaron - ----- Original Message ----- From: "Torben Schlüntz" To: Sent: Monday, October 29, 2001 3:23 PM Subject: SV: SV: Mersenne: number of processors participating > Yep! But the time entry only allows the program to sleep (still eating > all CPU cycles even when running at zero priority). Take any NT 4.0 or > W2K machine and you will see the "system idle time" doesn't add seconds > while Prime95 still eats them (and doing nothing). > For my servers to become prime95's I need to be sure they only run what > I have planned at anytime. > I can start Prime95 scheduled. I don't mind! > But the users should never have one chance of claiming servers aren't > available or even running "slow". > I know you are certain and I know you gotta be damn good at this (very > far beyond anything I will ever manage); but still any doubt will become > my users advantage. > Make the sleepy nights for my servers glorius. I make them start prime95 > by a schedule and You make prime95 die by harikiri - and I decide when > everything happens. :-) Tnx in advance. > > Still happy hunting > tsc > > > -----Oprindelig meddelelse----- > Fra: George Woltman > Sendt: ma 29-10-2001 22:47 > Til: Torben Schlüntz > Cc: > Emne: Re: SV: Mersenne: number of processors participating > > > > Hi, > > At 10:01 PM 10/29/2001 +0100, you wrote: > >I would like to use the servers; but I haven't been able to > persuade > >George to make a Quit function like > > > > quit_at: 06:00 > > > >to terminate the program when users arrives and optimum > performance is > >needed > > Look in readme.txt for the Time= entry in prime.ini > This feature can be used to make prime95 go dormant at a > specified time. > > Hope that helps, > George > > > > _________________________________________________________________________ > Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm > Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 20:34:23 -0500 From: Nathan Russell Subject: Re: SV: Mersenne: number of processors participating On Mon, 29 Oct 2001 17:05:02 -0800, "Aaron Blosser" wrote: > >Still the only time I've ever seen Prime95/NTPrime slow down a system is when I was doing some Netmeeting video conferences. > >With it running, the video conference would run DOG slow. Stop the NTPrime service and curiously had to restart the video conference for the effect, but the video would then be running great. > >That was with, umm.. version 20 I think? I haven't tried again with later versions... wasn't one of the things George did something to do with the priority setting? > >Aaron Out of curiousity, have you tried tinkering with the thread priorities of the programs in question? I find the utility bvslice (http://www.blueneptune.com/~maznliz/marius/software.shtml) to be quite useful. Nathan _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Mon, 29 Oct 2001 19:37:14 -0800 From: "John R Pierce" Subject: Re: SV: Mersenne: number of processors participating > Still the only time I've ever seen Prime95/NTPrime slow down a system is when I was doing some Netmeeting video conferences. > > With it running, the video conference would run DOG slow. Stop the NTPrime service and curiously had to restart the video conference for the effect, but the video would then be running great. > > That was with, umm.. version 20 I think? I haven't tried again with later versions... wasn't one of the things George did something to do with the priority setting? I've had similar problems with a few other multimedia sorts of junkware. Near as I can tell, some of these things put their video or animation thread at Idle_Priority+1 or something, and it gets eaten alive by Prime95. - -jrp _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 10:37:41 +0100 From: Lars Lindley Subject: Mersenne: Intel IA64 & AMD Sledgehammer Are there plans for making mprime and prime95 capable of using 64bit processors like IA64 and AMD Sledgehammer? I know that these processors are able to run 32 bit code but will there be a 64bit optimized version of mprime/prime95? Cheers /Lars _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 22:00:07 -0000 From: bjb@bbhvig.uklinux.net Subject: Re: Mersenne: Intel IA64 & AMD Sledgehammer On 30 Oct 2001, at 10:37, Lars Lindley wrote: > Are there plans for making mprime and prime95 capable of using 64bit > processors like IA64 and AMD Sledgehammer? > > I know that these processors are able to run 32 bit code but will > there be a 64bit optimized version of mprime/prime95? IA64 may run IA32 code but I seem to remember this is far from optimal. Hammer (which AFAIK exists only as vapourware) is supposed to run IA32 code efficiently. What would we expect to gain from extended addressing, or 64 bit integer operations? Probably not much for LL testing ... that's using double-precision floating-point, which is already native 64+ bit hardware in IA32. Trial factoring would almost certainly speed up a fair bit with 64 bit integer hardware operations. There may be mileage in further optimization for Itanium (IA64) so as to take better advantage of the parallel execution paths than the current "scalar" code. However Itanium processors are still a frightening price; I can't see them becoming "consumer items" in the near future. Also, Glucas already works very well on Itanium systems! Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 22:00:07 -0000 From: bjb@bbhvig.uklinux.net Subject: Re: SV: Mersenne: number of processors participating On 29 Oct 2001, at 19:37, John R Pierce wrote: > I've had similar problems with a few other multimedia sorts of > junkware. Near as I can tell, some of these things put their video or > animation thread at Idle_Priority+1 or something, and it gets eaten > alive by Prime95. Isn't it the old problem - no matter what priority a process is running at, unless it's interrupt driven it won't preempt a process running at a lower priority. The problem here is that the multimedia stuff wants to do a very little work but very often. It gets slowed down because Prime95 hangs on to the processor until its timeslice expires - it almost never has to wait for some external event. Ideally the multimedia stuff would be driven by timer interrupt. But for some reason (maybe something to do with there being a limited number of timer channels, and those having rather poor resolution) this approach seems to be quite rare on PC systems. One way to improve the performance in these circumstances is to reduce the minimum timeslice for low-priority processes. This will cause the task scheduler to be "busier" and therefore reduce the overall performance to some extent, but multimedia type applications will coexist much more happily with compute-intensive tasks if this is done. Sorry, I have no idea how to do this, or even whether it is possible, in any of the versions of Windows. The linux 2.4 kernel does this almost automatically, by having a much smaller minimum timeslice for idle-priority processes than for processes running above idle priority. (The timeslice is reduced again for processes running at unusually high priority, so that they can't hog the whole system quite so easily.) I believe the timeslice parameters are tunable (without having to recompile the kernel), but I have no personal experience of actually doing this. Another other way to "fix" the problem is to have the compute- intensive process voluntarily relinquish its timeslice at intervals which are much shorter than the minimum timeslice (which is typically of the order of 200 ms). This reduces the efficiency of the compute-intensive task to some extent but does make it coexist better. I suppose it would be possible to build this into Prime95; if this is done I would like options to be "multimedia friendly" or "optimally efficient" - probably the best way to implement would be to have the code contain the relevant system calls but to NOOP over them if efficiency is demanded. The remaining problem with this approach is that how often you would want to make these system calls would depend very heavily on the processor speed. Relinquishing the timeslice very frequently would enable even slow systems to run multimedia pretty seamlessly, but at a heavy cost on all systems. Placing the system calls in a position where they would be effective but not too costly, across a large range of processor speeds and a large range of FFT run lengths, would not be a trivial task. Regards Brian Beesley _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 23:28:41 +0100 From: "Steinar H. Gunderson" Subject: Mersenne: Re: number of processors participating On Tue, Oct 30, 2001 at 10:00:07PM -0000, bjb@bbhvig.uklinux.net wrote: >which are much shorter than the minimum timeslice (which is >typically of the order of 200 ms). 200ms? Wouldn't this be an error? I can't really imagine that one would typically have only five time slices per second :-) /* Steinar */ - -- Homepage: http://www.sesse.net/ _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 14:39:23 -0800 From: "John R Pierce" Subject: Re: SV: Mersenne: number of processors participating > > I've had similar problems with a few other multimedia sorts of > > junkware. Near as I can tell, some of these things put their video or > > animation thread at Idle_Priority+1 or something, and it gets eaten > > alive by Prime95. > > Isn't it the old problem - no matter what priority a process is > running at, unless it's interrupt driven it won't preempt a process > running at a lower priority. ... process and thread dispatching in MS Windows IS interrupt driven. Anything that causes a thread or process thats waiting to become ready will cause it to immediately dispatch if its the highest priority ready process, the system doesn't wait for the next major quantum tick. Multimedia stuff is either waiting on sound buffer events, or multimedia timer events (which have 1mS resolution) or disk IO buffer events, or software semaphore events, all of which are interrupt driven and will cause an immediate dispatch. near as I can guess, the issue here is that Prime95 is running a few priority notches above idle and when another process tries to run at a lower priority it will stall behind prime95. _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 14:42:55 -0800 From: "Aaron Blosser" Subject: RE: SV: Mersenne: number of processors participating > One way to improve the performance in these circumstances is to > reduce the minimum timeslice for low-priority processes. This will > cause the task scheduler to be "busier" and therefore reduce the > overall performance to some extent, but multimedia type > applications will coexist much more happily with compute-intensive > tasks if this is done. > > Sorry, I have no idea how to do this, or even whether it is possible, > in any of the versions of Windows. There is a program that can set the quanta for programs... let me find that durned thing... Aha. http://www.sysinternals.com/ntw2k/freeware/frob.shtml Good old sysinternals... they have the neatest tools. Apparently that's just for NT4 machines (I think...). For Win2K (and presumably XP?), they have another page that tells you about the settings on there, and to wait for a new version of Frob that works with win2k. http://www.sysinternals.com/ntw2k/info/nt5.shtml Aaron _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 23:55:13 +0100 From: =?utf-8?Q?Torben_Schl=C3=BCntz?= Subject: SV: SV: Mersenne: number of processors participating Ups, by help from Brian Beesley and a little work with the "time=" I have it working now. I think it was my old paranoia from a time when I was not running the servers alone - I wouldn't let anyone know that a program like prime95 was active. Now I don't care as I have nobody but users to face. Thanks to all. Happy hunting tsc -----Oprindelig meddelelse----- Fra: George Woltman Sendt: ma 29-10-2001 22:47 Til: Torben Schlüntz Cc: Emne: Re: SV: Mersenne: number of processors participating Hi, At 10:01 PM 10/29/2001 +0100, you wrote: >I would like to use the servers; but I haven't been able to persuade >George to make a Quit function like > > quit_at: 06:00 > >to terminate the program when users arrives and optimum performance is >needed Look in readme.txt for the Time= entry in prime.ini This feature can be used to make prime95 go dormant at a specified time. Hope that helps, George _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 21:33:59 -0500 From: Rick Pali Subject: RE: SV: Mersenne: number of processors participating Aaron Blosser wrote: >Good old sysinternals... they have the neatest tools. Damn straight! I've been using (and loving) PageDefrag since I stumbled on that site. A few other gems have since made their way onto my system... Rick. - -+--- rpali@alienshore.com http://www.alienshore.com/seeking/ _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 23:01:30 -0600 From: Ken Kriesel Subject: Re: Mersenne: Re: number of processors participating As I recall, 200 was the default quantum on a vax 780 20 years ago. Even 10 years ago, interactive response could be sped up a lot by cranking quantum down to single digits. At 11:28 PM 10/30/2001 +0100, Steinar H. Gunderson wrote: >On Tue, Oct 30, 2001 at 10:00:07PM -0000, bjb@bbhvig.uklinux.net wrote: >>which are much shorter than the minimum timeslice (which is >>typically of the order of 200 ms). > >200ms? Wouldn't this be an error? I can't really imagine that one would >typically have only five time slices per second :-) > >/* Steinar */ >-- >Homepage: http://www.sesse.net/ >_________________________________________________________________________ >Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm >Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers > _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Tue, 30 Oct 2001 23:01:38 -0600 From: Ken Kriesel Subject: Thrashing (was Re: SV: Mersenne: number of processors participating) What will really slow a workstation or server down is running short of RAM. These days the working sets are getting appreciable as the exponents increase. NT scheduling will wake up the service version of ntprime every second I think and give it at least one quantum. If some more essential service or application needs nearly all available RAM for its working set, and the working set of ntprime is big enough it gets paged out, the disk thrashes wildly and performance can suffer greatly for both the ntprime service and the other service or application, even while the ntprime service only gets a percent or two of cpu time. This is not just a characteristic of NT, but a general property of virtual memory operating systems; eventually it's just too little ram or too much demand, leading to performance decline. Ken At 05:05 PM 10/29/2001 -0800, Aaron Blosser wrote: >Still the only time I've ever seen Prime95/NTPrime slow down a system is when I was doing some Netmeeting video conferences. _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 31 Oct 2001 11:42:59 +0200 From: "Daidalos" Subject: Mersenne: Winter Time Zone change In this part of the neighborhood the conventional daily time has change since a few days, due to the coming of winter. Specifically, it has been set to one hour back. Of course, I want to update my computer's time, in order to agree with the outside world. If I stop Prime95, change the time, and then start Prime95 again, will then Prime95 start from the correct last point of the calculation, or will it start from the point which was saved one hour before? Do I realy have to stop the program before I change the time? If indeed there is a problem, one solution would be to change the time, and then wait one hour before starting Prime95 again. But I'm looking for a way to do this without loosing one hour. Saving an hour? Well, yeah, I'm a hard core, it seems. Lampros - -- Lampros Kallenos Lefkosia, Kypros Polytexnitis@hotmail.com . _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 31 Oct 2001 10:07:12 +0100 From: Lars Lindley Subject: Re: Mersenne: Winter Time Zone change You can change the time during calculations. The only thing that happens then is that you get a really big negative number as the calculation time on the next iteration. Then it's back to normal. If you turn prime95 off and on it will start where you stopped. It doesn't save the intermediate result based on time. It bases it on what iteration you are on. Keep up the fun! /Lars _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 31 Oct 2001 09:40:58 -0600 From: softmade@mailtag.com Subject: Mersenne: http://www.scruznet.com/~luke/signup.htm This link isn't working. _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Wed, 31 Oct 2001 16:53:41 +0100 From: Alexander Kruppa Subject: Re: SV: Mersenne: number of processors participating > Another other way to "fix" the problem is to have the compute- > intensive process voluntarily relinquish its timeslice at intervals > which are much shorter than the minimum timeslice (which is > typically of the order of 200 ms). This reduces the efficiency of the > compute-intensive task to some extent but does make it coexist > better. I suppose it would be possible to build this into Prime95; if > this is done I would like options to be "multimedia friendly" or > "optimally efficient" - probably the best way to implement would be > to have the code contain the relevant system calls but to NOOP > over them if efficiency is demanded. I've made good experience with throwing a sched_yield() into the MFAC code. The machines MFAC was running on had a Linux 2.4 kernel which gives even niceness 19 process about 10% cpu when another normal niceness process is running, which some users complained about. I wasn't particularly careful where I put the sched_yield(), I think it was called far more often than neccesary (many times/ms) but the effect on performance was not that dramatic - about 5% slowdown. The overhead in the scheduler seems to be pretty low. With that, when another process was running, MFAC worked at about 1/1000 the normal speed, so it must have gotten less than 0.1 % cpu time. Users were happy again. Perhaps the performance loss can be reduced by placing the yield somewhere in the code where the data currently in cache is finished and new data must be read from memory. If another process wants to run, at least it'll throw data out of the cache that we dont need anymore anyways (let some other job do our dirty flushing). Alex _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ Date: Thu, 01 Nov 2001 05:37:34 +1300 From: "Halliday, Ian" Subject: Mersenne: [Fwd: Luke Welsh's Email Address] This is a cryptographically signed message in MIME format. - --------------msCE04DE18AF975DA001C9083B Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Luke, the list owner, is in the process of changing hosting services. I haven't heard from him since he sent me the following, which suggests that the link is dead, but that he will be sorting it out. Regards, Ian - -------- Original Message -------- Subject: Luke Welsh's Email Address From: Luke Welsh Hello Everybody-- Well, it looks line my dear, old, original ISP is finally calling it quits. Sad. Soon my personal email address, Luke@SCruzNet.com, will cease to exist. I'll find a new home for my Mersenne pages, link-rot and all. But this may not happen for some time. I will try to get a larger disk quota so I can host the Mersenne-Digest mailing list archives. On the plus side, my spam-o-meter should register somewhat lower :-) - --------------msCE04DE18AF975DA001C9083B Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIIH8wYJKoZIhvcNAQcCoIIH5DCCB+ACAQExCzAJBgUrDgMCGgUAMAsGCSqGSIb3DQEHAaCC BcYwggKVMIIB/qADAgECAgMEoJ8wDQYJKoZIhvcNAQEEBQAwgZIxCzAJBgNVBAYTAlpBMRUw EwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93bjEPMA0GA1UEChMGVGhh d3RlMR0wGwYDVQQLExRDZXJ0aWZpY2F0ZSBTZXJ2aWNlczEoMCYGA1UEAxMfUGVyc29uYWwg RnJlZW1haWwgUlNBIDIwMDAuOC4zMDAeFw0wMTA0MTgwNzI1NDlaFw0wMjA0MTgwNzI1NDla MF0xETAPBgNVBAQTCEhhbGxpZGF5MQ4wDAYDVQQqEwVJYW4gVzEXMBUGA1UEAxMOSWFuIFcg SGFsbGlkYXkxHzAdBgkqhkiG9w0BCQEWEGlhbkBoYWxsaWRheS5jb20wgZ8wDQYJKoZIhvcN AQEBBQADgY0AMIGJAoGBAMK91g0/Rxl+a2diM6gVF6zL6JvzCjqctTeA9XLjqdGybXvcghhw Wzu9GIiq3QPTuw53SH5jLnh/El4M1BAP9E5gevH/lw0FgR7Vxg7Y4eTPgxoe3i9D1IT0Jb5d X5XAd4AdwCcPDqZnieYL8vgvCb2tpt01EsQXQJ8iNYck1KtBAgMBAAGjLTArMBsGA1UdEQQU MBKBEGlhbkBoYWxsaWRheS5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQCL UrXMblruY2VZsWs4ofEnTUHQ10qjMPKFb33YnGF1OHHgQOrZ6gpUZCpENOmduk6zzF9xndOf XMsWodSDCnuTmb8cvhoTN3ZoZ/LkYjsIQ8eCGCzvnfsDkZ8ogWSUqNUR1d/xr0oi3nxND2sG JedqZLGvgYj+6nUJpc6Y/nVIfzCCAykwggKSoAMCAQICAQwwDQYJKoZIhvcNAQEEBQAwgdEx CzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93 bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRpb24g U2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBD QTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0wMDA4 MzAwMDAwMDBaFw0wMjA4MjkyMzU5NTlaMIGSMQswCQYDVQQGEwJaQTEVMBMGA1UECBMMV2Vz dGVybiBDYXBlMRIwEAYDVQQHEwlDYXBlIFRvd24xDzANBgNVBAoTBlRoYXd0ZTEdMBsGA1UE CxMUQ2VydGlmaWNhdGUgU2VydmljZXMxKDAmBgNVBAMTH1BlcnNvbmFsIEZyZWVtYWlsIFJT QSAyMDAwLjguMzAwgZ8wDQYJKoZIhvcNAQEBBQADgY0AMIGJAoGBAN4zMqZjxwklRT7Sbngn Z4HF2ogZgpcO40QpimM1Km1wPPrcrvfudG8wvDOQf/k0caCjbZjxw0+iZdsN+kvx1t1hpfmF zVWaNRqdknWoJ67Ycvm6AvbXsJHeHOmr4BgDqHxDQlBRh4M88Dm0m1SKE4f/s5udSWYALQmJ 7JRr6aFpAgMBAAGjTjBMMCkGA1UdEQQiMCCkHjAcMRowGAYDVQQDExFQcml2YXRlTGFiZWwx LTI5NzASBgNVHRMBAf8ECDAGAQH/AgEAMAsGA1UdDwQEAwIBBjANBgkqhkiG9w0BAQQFAAOB gQBzG28mZYv/FTRLWWKK7US+ScfoDbuPuQ1qJipihB+4h2N0HG23zxpTkUvhzeY42e1Q9Dps NJKs5pKcbsEjAcIJp+9LrnLdBmf1UG8uWLi2C8FQV7XsHNfvF7bViJu3ooga7TlbOX00/LaW GCVNavSdxcORL6mWuAU8Uvzd6WIDSDGCAfUwggHxAgEBMIGaMIGSMQswCQYDVQQGEwJaQTEV MBMGA1UECBMMV2VzdGVybiBDYXBlMRIwEAYDVQQHEwlDYXBlIFRvd24xDzANBgNVBAoTBlRo YXd0ZTEdMBsGA1UECxMUQ2VydGlmaWNhdGUgU2VydmljZXMxKDAmBgNVBAMTH1BlcnNvbmFs IEZyZWVtYWlsIFJTQSAyMDAwLjguMzACAwSgnzAJBgUrDgMCGgUAoIGxMBgGCSqGSIb3DQEJ AzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTAxMTAzMTE2MzczNFowIwYJKoZIhvcN AQkEMRYEFFVuogcIal/HU9BWszM5f3jD+vEZMFIGCSqGSIb3DQEJDzFFMEMwCgYIKoZIhvcN AwcwDgYIKoZIhvcNAwICAgCAMAcGBSsOAwIHMA0GCCqGSIb3DQMCAgFAMA0GCCqGSIb3DQMC AgEoMA0GCSqGSIb3DQEBAQUABIGAgCRGFyLen1/zMYskkvEEnb3NZGPJdpBp3O3Y7PN/gFSG H/oQzW85koMeb38+31XPbMYtGOS9CwF4/vWO9XzKiy4EsbaiGfxgaLR7nABJoboJFfqC1ZGn fvDv5geDhjhS+FLlW1QCih3SR+s2sr/PR2JW8G7kL2vFMywZchmaym0= - --------------msCE04DE18AF975DA001C9083B-- _________________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers ------------------------------ End of Mersenne Digest V1 #898 ******************************