 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| HomeDiscussion GroupsSpace ScienceAstronomyAmateur AstronomySpace FlightSpace StationShuttleSpace HistorySpace PolicySETI SpaceKB.com Contact UsLink To UsSearch & Site Map |
Space Forum / SETI / September 2008SETI/BOINC Question.
Is there anybodyyyyy... Out there? <cue acoustic guitar> I've recently returned to crunching a few units again after an hiatus as I have use of a wattage meter and discovered that running SETI/BOINC on a 30% duty cycle only consumes an extra 10W on my C2D @ 3.3GHz. The question: I'm running Seti@home Enhanced client 5.27 which is a beta client. When I check the "Your Results" link I see around ~100 tasks that say "In Progress" when I only have <10 queued up in the client. Can anyone tell me what's going on? TIA. Also, I see that the client has just now downloaded the Astropulse .exe and a 10MB Astropulse WU that is estimated just a bit under 24 hours to completion. I've never done an Astropulse WU before. Being ~24 hours to completion and running a 30% duty cycle it'll take over three days to calculate. I'm pleased the reporting date is a month away. I only hope that I get credit for it, that it isn't done and dusted by the time I report it. Regards,  SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) > Is there anybodyyyyy... Out there? <cue acoustic guitar> > [quoted text clipped - 6 lines] > say "In Progress" when I only have <10 queued up in the client. Can anyone > tell me what's going on? TIA. You've picked up what are called 'ghost WU's'. There have been glitches on the SETI servers where it thought that it had sent WU's out, but the client's computer didn't get them. Just ignore them and they will go away. > Also, I see that the client has just now downloaded the Astropulse .exe and > a 10MB Astropulse WU that is estimated just a bit under 24 hours to > completion. I've never done an Astropulse WU before. Being ~24 hours to > completion and running a 30% duty cycle it'll take over three days to > calculate. I'm pleased the reporting date is a month away. I only hope that > I get credit for it, that it isn't done and dusted by the time I report it. I've gotten 2 Astorpulse WU's, so far, on my 3.0 ghz P4/HT, and on a laptop. They will be taking a few days to complete. Just let it run. :-) > Regards, SignatureClaude Somewhere on teh intarweb "Claude Ortega" typed: >> Is there anybodyyyyy... Out there? <cue acoustic guitar> >> [quoted text clipped - 12 lines] > the client's computer didn't get them. Just ignore them and they > will go away. Ahh, thanks for that. I'm pleased that it's a known problem. I was concerned thatit could be something specific to me/my computer. The computer summary page lists 247 tasks for my computer! >> Also, I see that the client has just now downloaded the Astropulse >> .exe and a 10MB Astropulse WU that is estimated just a bit under 24 [quoted text clipped - 6 lines] > I've gotten 2 Astorpulse WU's, so far, on my 3.0 ghz P4/HT, and on a > laptop. They will be taking a few days to complete. Just let it run. :-) Thanks again. That one's done (26,730.88 secs of CPU time, 168.62 claimed credit, pending) and I see I have another one queued. I looked at the SETI@home site after posting this (yeah, I know..) and see that there are requirements that must be met by the computer before it's sent Astropulse units. Obviously the alogrithm that decides these things likes my computer. (Measured floating point speed 3215.65 million ops/sec, Measured integer speed 7047.89 million ops/sec per CPU.) Thanks for the reply Claude. SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Claude Ortega" typed: >> I've gotten 2 Astorpulse WU's, so far, on my 3.0 ghz P4/HT, and on a >> laptop. They will be taking a few days to complete. Just let it run. :-) [quoted text clipped - 6 lines] >(Measured floating point speed 3215.65 million ops/sec, Measured integer >speed 7047.89 million ops/sec per CPU.) So how do you recognize the Astropulse WUs? I have seen WUs which use/need client version 5.28 to process (where it used to be 5.27), but nothing specific about 'Astropulse'. (Keep in mind I am running optimized Set clients, so perhaps BOINC doesn't accept/request Astropulse units because of that.) On another note: Anyone tried installing the new 6.2.14 (or something like that) of BOINC? I installed it without problems on my Linux machines, but when installing it on a Windows box, I noticed a new question in the dialog box: it wants two directories now: 1 for BOINC itself, 1 for data. However, the defaults do not place the 'Data' directory inside the BOINC directory, but somewhere else. And, when setting the data directory to the BOINC directory, the installer complains one cannot set these the same (they were identical on 5.10.45!). Fearing I might lose all WUs in progress, I haven't installed it yet. Anyone know the deal on this? What should I pick, and can I move the Seti@Home WUs across to the new dir? Or should I do something else first? Regards, Patrick. Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Somewhere on teh intarweb "Claude Ortega" typed: > [quoted text clipped - 10 lines] >> speed 3215.65 million ops/sec, Measured integer speed 7047.89 >> million ops/sec per CPU.) Hi Patrick. > So how do you recognize the Astropulse WUs? I have seen WUs which > use/need client version 5.28 to process (where it used to be 5.27), > but nothing specific about 'Astropulse'. Astropulse WUs are prefixed by "ap_", have lots longer file names and have run times listed that are approximately 5 x that of 'normal' WUs. Also, under the "Application" column in Task Manager it shows Astropulse 3.45 instead of SETI@home Enhanced 5.27. (Astroplulse 3.45.exe is downloaded when you get the first Astropulse WU.) > (Keep in mind I am running optimized Set clients, so perhaps BOINC > doesn't accept/request Astropulse units because of that.) Hmmm, I remember seeing something that might have been about that on the Astropulse page (linked from SETI@home home page). I think you have to download a new file? From: http://setiathome.berkeley.edu/ap_faq.php "If you are using an optimised application then you will need to replace or modify the app_info.xml file. An app_info.xml file will be made available here." However, I don't think that the file is there yet. I don't see a link to it. > On another note: Anyone tried installing the new 6.2.14 (or something > like that) of BOINC? I'm using 6.2.12 which was listed as 'beta' when I downloaded it a couple weeks ago. I've had no problems with it. > I installed it without problems on my Linux > machines, but when installing it on a Windows box, I noticed a new [quoted text clipped - 5 lines] > installer complains one cannot set these the same (they were > identical on 5.10.45!). I set my BOINC directory to my programmes partition and the data directory to my data drive. I liked having the option, it means my programmes partition doesn't get too fragmented over time with BOINC constantly writing/deleting data. > Fearing I might lose all WUs in progress, I haven't installed it yet. > Anyone know the deal on this? What should I pick, and can I move the > Seti@Home WUs across to the new dir? Or should I do something else > first? Hard to know for sure, I installed 6.2.12 after not running BOINC for a while. However, I seem to remember seeing a couple of long-expired WUs in BOINC task manager when I started it so maybe it moved them. If I were you I'd set my preferences to 0 / 0 days a while before you install it, just in case. Even worst case scenario is only losing the WUs that the client is currently working on as downloaded but unprocessed units haven't actually cost you any CPU time.. Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Patrick Vervoorn" typed: >Hi Patrick. Hey Shaun, >> So how do you recognize the Astropulse WUs? I have seen WUs which >> use/need client version 5.28 to process (where it used to be 5.27), [quoted text clipped - 5 lines] >instead of SETI@home Enhanced 5.27. (Astroplulse 3.45.exe is downloaded when >you get the first Astropulse WU.) Aha, haven't spotted anything like that, even on machines which are clearly capable of processing these WUs. >> (Keep in mind I am running optimized Set clients, so perhaps BOINC >> doesn't accept/request Astropulse units because of that.) [quoted text clipped - 10 lines] > >However, I don't think that the file is there yet. I don't see a link to it. Nope, I don't see anything either. Also nothing from the site I downloaded these optimized binaries from (the 'lunatics' site from the third-party list). >> On another note: Anyone tried installing the new 6.2.14 (or something >> like that) of BOINC? > >I'm using 6.2.12 which was listed as 'beta' when I downloaded it a couple >weeks ago. I've had no problems with it. I did install 6.3.x on a Linux machine of mine a while ago, but that was beta then. >> I installed it without problems on my Linux >> machines, but when installing it on a Windows box, I noticed a new [quoted text clipped - 10 lines] >partition doesn't get too fragmented over time with BOINC constantly >writing/deleting data. I have programs and data mixed. ;) >> Fearing I might lose all WUs in progress, I haven't installed it yet. >> Anyone know the deal on this? What should I pick, and can I move the [quoted text clipped - 8 lines] >currently working on as downloaded but unprocessed units haven't actually >cost you any CPU time.. I'll probably backup the directory, and just try it on one of my slower Windows-based machines to see what happens. I must say I find this change in the installer a bit confusing at best; a notice about the current location of the data, and perhaps explicitly having it mention that data will or will not be moved, would've been nice... Regards, Patrick. Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Somewhere on teh intarweb "Patrick Vervoorn" typed: [selective snippage] >> Hi Patrick. > > Hey Shaun, Good to 'see' you mate. >> Astropulse WUs are prefixed by "ap_", have lots longer file names >> and have run times listed that are approximately 5 x that of [quoted text clipped - 5 lines] > Aha, haven't spotted anything like that, even on machines which are > clearly capable of processing these WUs. It seems that they've only been rolled out in the last couple weeks. >> From: http://setiathome.berkeley.edu/ap_faq.php >> [quoted text clipped - 8 lines] > downloaded these optimized binaries from (the 'lunatics' site from > the third-party list). I didn't realise how new Astropulse is, seems I came back to SETI just as they introduced it. >> I set my BOINC directory to my programmes partition and the data >> directory to my data drive. I liked having the option, it means my >> programmes partition doesn't get too fragmented over time with BOINC >> constantly writing/deleting data. > > I have programs and data mixed. ;) I'm a neat freak. (I prefer 'optimised' really.) I have multiple HDDs in my main machine so like to split I/O between disks but even in my PCs that have a single disk I have seperate partitions for OS, programmes and data. I find that systems run faster like that and it's easier to keep fragmentation to a minimum. >> Hard to know for sure, I installed 6.2.12 after not running BOINC >> for a while. However, I seem to remember seeing a couple of [quoted text clipped - 10 lines] > the current location of the data, and perhaps explicitly having it > mention that data will or will not be moved, would've been nice... It seems that Tazz has experience with this so it might be a good idea to flush your cache first. Perhaps now? While the servers are down your cache might be empty? I'm just glad that I have a couple Astropulse WUs queued (last one processing now) so that I had work during this server outage. In a couple hours my cores are going to go into rest mode. Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Patrick Vervoorn" typed: >[selective snippage] Good idea, I'll do the same, where applicable of course. ;) >> Hey Shaun, > >Good to 'see' you mate. Yo. ;) >>> Astropulse WUs are prefixed by "ap_", have lots longer file names >>> and have run times listed that are approximately 5 x that of [quoted text clipped - 7 lines] > >It seems that they've only been rolled out in the last couple weeks. And it seems they only end up on machines which have an unchanged/non-existing app_info.xml, or a modified app_info.xml. Checked my machines, and a P3-700MHz, 512MB Linux machine of mine, for which I haven't found any optimized binaries, has one queued up, expected run-time: ~205 hours. Weird, since I don't meet the 1.6GHz requirement, but we'll see how this turns out. >> Nope, I don't see anything either. Also nothing from the site I >> downloaded these optimized binaries from (the 'lunatics' site from >> the third-party list). > >I didn't realise how new Astropulse is, seems I came back to SETI just as >they introduced it. It's something from the last few weeks or so, although there have apparently been rumblings about it for much, much longer. >> I have programs and data mixed. ;) > [quoted text clipped - 3 lines] >that systems run faster like that and it's easier to keep fragmentation to a >minimum. I did that, even partitioned disks in a primary C: for the OS, a D: for Programs and an E: for data, but as time went on, I noticed I was running out of space on all of them, so nowadays I stick to a somewhat larger C: (25GB for XP/Vista32 seems to be fine, Vista64 needs a bit more, so I set that at 50GB), and the rest of the disk as one big D: >> I'll probably backup the directory, and just try it on one of my >> slower Windows-based machines to see what happens. I must say I find [quoted text clipped - 7 lines] >(last one processing now) so that I had work during this server outage. In a >couple hours my cores are going to go into rest mode. Well, I've set my preferences to 5+5 days of work, so all machines are still crunching on, and aren't even halfway past their cache. Also I think BOINC re-downloads any missing WUs when you, for instance, wipe the directory. My 'plan' (still not done, since I noticed they juggled the versions around a bit: one day they offer 6.2.14, then 5.10.45 again, and now they are again offering 6.2.15 as the new BOINC versions) is to backup the directory, then install the new BOINC and monitor the stuff. Also, perhaps the 6.2.15 windows installer works differently from the previous 6.2.14 (beta) which was, so it seems, mistakenly launched as a final.... I also ran a 6.3.6 beta client for a while, and one of the most interesting messages that spat out during startup was the line "No CUDA devices found". I suppose it would be very neat if I could use my 8800GTX as a Seti@Home cruncher, but no client is in sight AFAIK. Darn. Ciao, Patrick. Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Somewhere on teh intarweb "Patrick Vervoorn" typed: > [quoted text clipped - 27 lines] > expected run-time: ~205 hours. Weird, since I don't meet the 1.6GHz > requirement, but we'll see how this turns out. That is odd indeed. ~205 hours huh? Just for the hell of it, when I got my first couple, I checked to see what other PCs they were sent out to and the others all met the requirements although they were all heaps slower than mine. My 'average work done' has suffered since I started getting Astropulse units as they're taking a long time to verify. (None has yet..) >>> Nope, I don't see anything either. Also nothing from the site I >>> downloaded these optimized binaries from (the 'lunatics' site from [quoted text clipped - 5 lines] > It's something from the last few weeks or so, although there have > apparently been rumblings about it for much, much longer. Yes. I stopped crunching about 6 months ago for a while as my electricity bill is a bit much to manage. However, I have a meter that measures wattage consumed in real time and, when it was hooked up to my machine I found that, running SETI at a 50% duty cycle, I was only using ~20W more so decided that I could afford that for a while. They were talking about Astropulse back before my hiatus. >>> I have programs and data mixed. ;) >> [quoted text clipped - 10 lines] > needs a bit more, so I set that at 50GB), and the rest of the disk as > one big D: Ok. I have XP Pro on a 5GB partition which it shares with what I class as 'OS-essential" apps such as AV, Acronis, PerfectDisk, PartitionMagic, Spybot... There's still 1GB free after a year on this install. I don't have my pagefile in that partition though. Programmes is 15GB and is nearly full, I might have to increase it a bit with PartitionMagic. I have a 2GB FAT32 partition for pagefile only, at the start of a fast disk. The balance of those disks are large data partitions and the other disks are one big partition also. I find that, as I keep multiple backups (Acronis True Image) of the boot partition and sometimes the Programmes partition keeping them small is good. (Although Acronis truncates the free space anyway so I'm not sure that backups would be *that* much larger on bigger partitions with the same amount of data.) >>> I'll probably backup the directory, and just try it on one of my >>> slower Windows-based machines to see what happens. I must say I find [quoted text clipped - 11 lines] > Well, I've set my preferences to 5+5 days of work, so all machines are > still crunching on, and aren't even halfway past their cache. Good move. Mine is 1+1 and I was out of work for a while, then got a couple units last night, now am out of work again. :-( I'm waiting on delivery of a 45nm CPU, an E7300 that should arrive any minute. My E4500 has been running at 3.32GHz for a while now, time for a new challenge. The E7300 has a 10x multiplier so I'm hoping that it will run on a 400MHz FSB (1600 in Intel-speak) for 4GHz at sane vcore and temps. <fingers crossed> It's actually quite a big job as my cooler has a "bolt-thru" kit fitted which means the mobo has to come out and the case is quite cramped with all the HDDs and a full-length GFX card, SCSI card, sound card..... Also, I lapped the E4500 which gave me ~6°C cooler temps so I might have to fit the E7300 twice. (I don't want to lap it out-of-the-box until I'm sure it'll at least work!) > Also I think BOINC re-downloads any missing WUs when you, for > instance, wipe the directory. My 'plan' (still not done, since I [quoted text clipped - 4 lines] > installer works differently from the previous 6.2.14 (beta) which > was, so it seems, mistakenly launched as a final.... Ok. > I also ran a 6.3.6 beta client for a while, and one of the most > interesting messages that spat out during startup was the line "No > CUDA devices found". I suppose it would be very neat if I could use > my 8800GTX as a Seti@Home cruncher, but no client is in sight AFAIK. > Darn. Yeah. There's a client for ATI cards no? (I have nVidia too) I can imagine that running on both CPU and GPU would bump up the power consumption quite a bit! Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Checked my machines, and a P3-700MHz, 512MB Linux machine of mine, for >> which I haven't found any optimized binaries, has one queued up, >> expected run-time: ~205 hours. Weird, since I don't meet the 1.6GHz >> requirement, but we'll see how this turns out. > >That is odd indeed. ~205 hours huh? That's what the Linux utility 'boinc_curses' (a Linux console only clone of the BOINC Manager) claims/shows. >Just for the hell of it, when I got my >first couple, I checked to see what other PCs they were sent out to and the >others all met the requirements although they were all heaps slower than >mine. My 'average work done' has suffered since I started getting Astropulse >units as they're taking a long time to verify. (None has yet..) My RAC has more or less doubled on the Q6600 machine after I installed the AK8 SSSE3.1 binaries. ;) I'm not yet clear on how I have to merg the app_info.xml files to get boinc to use the 'regular' astropulse client for ap units, and the optimized stuff for the regular units. >> I did that, even partitioned disks in a primary C: for the OS, a D: >> for Programs and an E: for data, but as time went on, I noticed I was [quoted text clipped - 7 lines] >Spybot... There's still 1GB free after a year on this install. I don't have >my pagefile in that partition though. Hmmm, I did that (5GB C:), but that gets small very quickly. That is/was with swap on C: too, so that then gets moved to D:. >Programmes is 15GB and is nearly full, I might have to increase it a bit >with PartitionMagic. I have a 2GB FAT32 partition for pagefile only, at the [quoted text clipped - 4 lines] >truncates the free space anyway so I'm not sure that backups would be *that* >much larger on bigger partitions with the same amount of data.) I don't make too many images; a re-install is done quick enough, and the only sure way to get a clean machine again. ;) >> Well, I've set my preferences to 5+5 days of work, so all machines are >> still crunching on, and aren't even halfway past their cache. > >Good move. Mine is 1+1 and I was out of work for a while, then got a couple >units last night, now am out of work again. :-( I had it set a lot more 'conservative' at first, but after the last big outages (I think these happened somewhere in 2007), where the servers were out for days, I just set it all to the maximum, more or less after those events, when all my machines 'ran dry'. They haven't run dry since then. Yet anyhow.... >I'm waiting on delivery of a 45nm CPU, an E7300 that should arrive any >minute. My E4500 has been running at 3.32GHz for a while now, time for a new [quoted text clipped - 6 lines] >might have to fit the E7300 twice. (I don't want to lap it out-of-the-box >until I'm sure it'll at least work!) Sounds like a nice project. ;) Good luck taking that thing to the limit! >> I also ran a 6.3.6 beta client for a while, and one of the most >> interesting messages that spat out during startup was the line "No [quoted text clipped - 5 lines] >that running on both CPU and GPU would bump up the power consumption quite a >bit! Not that i know. I suppose an 8800GTX running at full-bore would be quite an energy-slurper. ;) No idea how much Seti@Home tasks it could, in theory, run. I find some snippets here and there, but no client yet. We'll see if I can harness the power of the GPU in the future... ;) Ciao, Patrick. <snip> > My RAC has more or less doubled on the Q6600 machine after I installed the > AK8 SSSE3.1 binaries. ;) I'm not yet clear on how I have to merg the > app_info.xml files to get boinc to use the 'regular' astropulse client for > ap units, and the optimized stuff for the regular units. All you have to do is open the 'sample' app_info.xml file with wordpad or notepad and highlight everything from the top down to and including </app_version>. Then copy it. Then, on your 'real' app_info.xml, open it in wordpad or notepad and highlight <app_info>. Then paste over it. When your done it should look like the 'sample' but with your optimized client info under the Astropulse suff. And remember to put the AP files that you downloaded in the setiathome.berkeley.edu folder. Restart BOINC and you should be good to go. <snip> Signature</Tazz> Somewhere on teh intarweb "Patrick Vervoorn" typed: [snippety-snip] >> Ok. I have XP Pro on a 5GB partition which it shares with what I >> class as 'OS-essential" apps such as AV, Acronis, PerfectDisk, [quoted text clipped - 3 lines] > Hmmm, I did that (5GB C:), but that gets small very quickly. That > is/was with swap on C: too, so that then gets moved to D:. That's odd. My C: is fine at 5GB, even with XP SP3 installed (uninstall files located on data partition). It still has ~1GB free despite me having a few 3rd party apps/utils installed there (Java, AVG, Acronis, PerfectDisk, Nero...). >> Programmes is 15GB and is nearly full, I might have to increase it a >> bit with PartitionMagic. I have a 2GB FAT32 partition for pagefile [quoted text clipped - 8 lines] > I don't make too many images; a re-install is done quick enough, and > the only sure way to get a clean machine again. ;) Heh! I have a heavilly customised interface and it's not that quick to reinstall for me. An image is much, much quicker. >> I'm waiting on delivery of a 45nm CPU, an E7300 that should arrive >> any minute. My E4500 has been running at 3.32GHz for a while now, [quoted text clipped - 10 lines] > Sounds like a nice project. ;) Good luck taking that thing to the > limit! Well, I'm a bit disappointed with the new CPU. It's only doing 3.6GHz stably at sane vcore (<1.4V). I say "only" as the 65nm E4500 was doing 3.32 at <1.4V. Still, the slight speed increase coupled with the 50% bigger L2 and new instruction-sets must help I suppose. Floating point and integer MIPs have gone up to 3750 and 7750 respectively. (~600 increase in each.) I've got a few completed Astropulse WUs now (and another queued) but still none that have been confirmed, all still pending. I see that they only send out two copies of each (last I remember it was three of each for vanilla SETI WUs?) with one of the WUs I'm waiting on having been sent to a Pentium III 1133MHz machine that does 1068 / 1763 MIPs respectively. I could be waiting an age for that one to be verified. Another, that took my machine 34 hours to crunch, has been sent to a Pentium Dual core 3.2GHz, 1457 / 2610 MIPs machine. That's about 1/3 the speed of my machine so ~100 hours crunch time on that machine? I reported it 8 days ago and it's still pending.... >> Yeah. There's a client for ATI cards no? (I have nVidia too) I can >> imagine that running on both CPU and GPU would bump up the power [quoted text clipped - 5 lines] > client yet. We'll see if I can harness the power of the GPU in the > future... ;) I remember seeing somethig about GPUs being used for BOINC... A quick Google gave me a couple /. articles... http://hardware.slashdot.org/article.pl?sid=06/10/13/2030253 http://tech.slashdot.org/article.pl?sid=08/07/19/1543204&from=rss But it is /. after all.... Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) Somewhere on teh intarweb "~misfit~" typed: > Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Hmmm, I did that (5GB C:), but that gets small very quickly. That [quoted text clipped - 4 lines] > despite me having a few 3rd party apps/utils installed there (Java, > AVG, Acronis, PerfectDisk, Nero...). Should have mentioned that SOP for me is to not only put the pagefile on another drive but also redirect 'My Documents", OE store and IE cache to a data drive. (Not that I use IE...) Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "~misfit~" typed: >> Somewhere on teh intarweb "Patrick Vervoorn" typed: [quoted text clipped - 9 lines] >another drive but also redirect 'My Documents", OE store and IE cache to a >data drive. (Not that I use IE...) I don't use the 'My Documents' and also don't use Outlook Express. The cache remains on C: though, as does the inital pagefile. Things tend to get a bit cluttered after about half a year to a year that way with a 5GB C:, which is why I upped it on my recent machines. Also, with the current size of HDs, 5 or 10 or 25GB doesn't make much difference anymore. ;) Regards, Patrick. Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Somewhere on teh intarweb "~misfit~" typed: >>> Somewhere on teh intarweb "Patrick Vervoorn" typed: [quoted text clipped - 11 lines] > > I don't use the 'My Documents' and also don't use Outlook Express. I use 'My Documents' as quite a few programmes want to save there by default. I found it easier to make one change (the location of 'My Documents' rather than chamge each programmes default save path. I use OE due to inertia. It's what I started using and, even though I've tried Agent, Mozilla and a few others I find them all a bit of a hassle. I just want to get the posts are reply to them in the manner to which I've become accustomed. Every other newsreader has been slower for me to use or doesn't thread properly.... > The > cache remains on C: though, as does the inital pagefile. I much prefer a fixed-size pagefile on a seperate partition, preferably a seperate HDD. > Things tend > to get a bit cluttered after about half a year to a year that way > with a 5GB C:, which is why I upped it on my recent machines. Fair enough. > Also, with the current size of HDs, 5 or 10 or 25GB doesn't make much > difference anymore. ;) I find that Windows is a lot faster (noticeably so) when it's loading and working from a small area of disk. When confined like that it can't throw temp files all over the disk, then thrash the heads when it needs to access them. <shrug> Whatever works huh? ;-) SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Patrick Vervoorn" typed: >> I don't use the 'My Documents' and also don't use Outlook Express. > >I use 'My Documents' as quite a few programmes want to save there by >default. I found it easier to make one change (the location of 'My >Documents' rather than chamge each programmes default save path. I think I've taught the programs who want to do this 'by default', to not do this. ;) >I use OE due to inertia. It's what I started using and, even though I've >tried Agent, Mozilla and a few others I find them all a bit of a hassle. I >just want to get the posts are reply to them in the manner to which I've >become accustomed. Every other newsreader has been slower for me to use or >doesn't thread properly.... I actually started reading news with trn, and still am. Nothing threads like trn does. Nothing is faster for browsing through a newsgroup than trn. But trn is not for everyone. And is also quite old. And quite not-maintained anymore, unfortunately. But, since I have the (already lightly modified) source code available, I can always (try to) fix problems with it myself. >I much prefer a fixed-size pagefile on a seperate partition, preferably a >seperate HDD. I've done that a few times, but never noticed any speedup. Of course, one should avoid use of the pagefile at all. ;) >> Also, with the current size of HDs, 5 or 10 or 25GB doesn't make much >> difference anymore. ;) [quoted text clipped - 3 lines] >temp files all over the disk, then thrash the heads when it needs to access >them. <shrug> Whatever works huh? ;-) That how I do it, but I'm open to other people's suggestions/tips of course. If/when I need to (re-)install another XP machine, I'll keep the info from this thread in mind... Regards, Patrick. Somewhere on teh intarweb "~misfit~" typed: > I've got a few completed Astropulse WUs now (and another queued) but > still none that have been confirmed, all still pending. I see that [quoted text clipped - 3 lines] > 1763 MIPs respectively. I could be waiting an age for that one to be > verified. I've been checking the results of my Astropulse WUs (just reported another one) and this guy let it run for quite a while before aborting it: http://setiathome.berkeley.edu/workunit.php?wuid=309983638 Luckilly it was given to another machine (Athlon 64 3000+) that was running 24/7 and it's now been completed by two machines so I've been credited for it. > Another, that took my machine 34 hours to crunch, has been sent to a > Pentium Dual core 3.2GHz, 1457 / 2610 MIPs machine. That's about 1/3 > the speed of my machine so ~100 hours crunch time on that machine? I > reported it 8 days ago and it's still pending.... http://setiathome.berkeley.edu/workunit.php?wuid=310230714 Still waiting on this one, I reported it 11 days ago. However, it was a big unit. The one I reported today was also a big one: http://setiathome.berkeley.edu/workunit.php?wuid=315852694 was also sent to a P4 2.4GHz, I hope he doesn't let it run for a while then abort too. SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "~misfit~" typed: >> I've got a few completed Astropulse WUs now (and another queued) but [quoted text clipped - 13 lines] >24/7 and it's now been completed by two machines so I've been credited for >it. After clearing the cache of regular WUs, my P3-700 Linux machine is finally crunching the lone Astropulse WU it got. It has currently spent about 71 hours on it, and BOINC estimates it will take another 227 hours to finish it. It's also currently the only work on this machine, I suppose it'll start requesting more once the AP unit is close to finishing... I do hope they don't send an Astropulse unit by accident to the P1-133MHz/64MB machine I also still have crunching (well, nibbling would be a more appropriate term perhaps ;) away. It takes about 225 hours just to cruch a regular WU. ;) Regards, Patrick. Somewhere on teh intarweb "Patrick Vervoorn" typed: >> Somewhere on teh intarweb "~misfit~" typed: >>> I've got a few completed Astropulse WUs now (and another queued) but [quoted text clipped - 21 lines] > machine, I suppose it'll start requesting more once the AP unit is > close to finishing... I've done quite a few now, my RAC really see-saws while I'm waiting for a few to be validated. I sometimes check to see the progress of the WU and quite a few are being aborted, I assume by the user. Also I'm seeing a higher than usual amount of 'compute errors' from other machines with Astropulse WUs. I think that SETI need to raise the bar as far as minimum requirements go. My current AP WU is the 'biggest' by far, having already taken 37 hours and the client is predicting another 10 hours. That's just one core working on it but that core is benched by SETI at 3500 / 7200. I see these units going to machines that are rated at 1000 / 1800. They'd take a couple weeks to crunch, especially if they're not running 24/7. What MIPS does BOINC give for that P3-700 Patrick? > I do hope they don't send an Astropulse unit by accident to the > P1-133MHz/64MB machine I also still have crunching (well, nibbling > would be a more appropriate term perhaps ;) away. It takes about 225 > hours just to cruch a regular WU. ;) LMAO! As these WUs look to expire after one month I really hope they raise the bar. I'm yet to have a AP WU validated when reported, they all take varying amounts of time to get a second result, sometimes weeks. Cheers, SignatureShaun. DISCLAIMER: If you find a posting or message from me offensive, inappropriate, or disruptive, please ignore it. If you don't know how to ignore a posting, complain to me and I will be only too happy to demonstrate... ;-) >Somewhere on teh intarweb "Patrick Vervoorn" typed: >> After clearing the cache of regular WUs, my P3-700 Linux machine is >> finally crunching the lone Astropulse WU it got. It has currently [quoted text clipped - 8 lines] >higher than usual amount of 'compute errors' from other machines with >Astropulse WUs. Since all my fast machines are still running a modified main SetBOINC app by default, they don't get any AP WUs, so I have little to compare against. I did fire up the secondary VistaPro64 install on the Q6600 machine (try-out install to see how Vista64 performs) about half a week ago, and since I let that one run out of WUs a long while ago, decided to install a version 6.x BOINC client on that. During the install of that, I did notice a (very quick) notice flashing by which claimed something like 'migrating data to ....'. So it _looks_ like that BOINC 6.x installer does possibly include the moving/copying of the data to the new data-directory defined in the installer. However, I'm still flushing the caches on my primary machines, so I'll probably try another install with a few WUs left once I can do that... Taking no risks here. ;) >I think that SETI need to raise the bar as far as minimum requirements go. Yeah. See below though for the 'why' of my 'yeah'. >My current AP WU is the 'biggest' by far, having already taken 37 hours and >the client is predicting another 10 hours. That's just one core working on [quoted text clipped - 3 lines] > >What MIPS does BOINC give for that P3-700 Patrick? The latest benchmark run on the P3-700 running Linux gave it 410 floating point MIPS and 718 integer MIPS. It's currently still crunching the AP WU it got quite a while ago. Used CPU time is, as I type this: 395:41:15. BOINC predicts it will take another 57:45:45 hours to complete. The report deadline is at 08-Sep-08 23:35:52, so it looks like the machine will _just_ about make it. So yeah, if/when they send out these suckers, please take good notice of the speed-rating of the machine you're sending it to, and also please perhaps extend the deadline a bit? >> I do hope they don't send an Astropulse unit by accident to the >> P1-133MHz/64MB machine I also still have crunching (well, nibbling [quoted text clipped - 4 lines] >the bar. I'm yet to have a AP WU validated when reported, they all take >varying amounts of time to get a second result, sometimes weeks. Some extension on regular WUs would also be nice; I think at least one, perhaps more, WUs crunched on the P1-133 have gone over the deadline. Especially since that little bugger had a habit of crashing regularly, before I finally found a kernel that would remain stable on it. In case anyone is wondering/interested: frodo:~$ uptime 20:13:23 up 16 days, 5:24, 1 user, load average: 1.21, 1.06, 1.02 frodo:~$ uname -a Linux frodo 2.6.18-4-486 #1 Wed May 9 22:23:40 UTC 2007 i586 GNU/Linux This is a Debian-stable kernel, versions 2.6.18-5-486 and up had a habit of crashing after only a few hours, up to at most a few days, of running, so I stuck to 2.6.18-4-486. This for people thinking about keeping an old Pentium 1 133MHz machine running and even doing something useful. ;) Regards, Patrick. >During the install of that, I did notice a (very quick) notice flashing by >which claimed something like 'migrating data to ....'. So it _looks_ like [quoted text clipped - 3 lines] >another install with a few WUs left once I can do that... Taking no risks >here. ;) Update: I finally flushed the caches on my primary Q6600 machines. I then one-time enabled it to get more work, so it fetched 20 WU's. I then quit BOINC, installed version 6.2.18, and noticed that same 'moving application data' messages. After starting BOINC 6.2.18, I noticed it still had those 20 WU's, and was also still processing them with the optimzed App. So... It seems no WUs are lost when upgrading from 5.x to 6.x.... I'll be upgrading my other machines during this weekend then... Regards, Patrick. <snip> > It seems that Tazz has experience with this so it might be a good idea to > flush your cache first. Perhaps now? While the servers are down your cache > might be empty? I'm just glad that I have a couple Astropulse WUs queued > (last one processing now) so that I had work during this server outage. In a > couple hours my cores are going to go into rest mode. With an un-modified app_info.xml I got 3 AP WUs that I couldn't process. Since I edited it, not a one! I copied my 5x BOINC folder before I installed 6.2.14. When I realized what was going on I disabled network access before any WUs were lined up for download. I went back to 5x and finished up the work there and installed 6x again. In the past I've never had BOINC re-download a WU. They stay in my tasks (on the website) until they run past the deadline. But, it's been quite a while since I've lost a WU and things may have changed. I pretty much backup before I change anything. </Tazz> >> Somewhere on teh intarweb "Claude Ortega" typed: > [quoted text clipped - 14 lines] > (Keep in mind I am running optimized Set clients, so perhaps BOINC doesn't > accept/request Astropulse units because of that.) You can manually download the Astropulse files and edit your app_info file to be able to process the AP WUs. http://setiathome.berkeley.edu/ap_advanced.php http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=1363&nowrap=true#34336 Scroll down a little for instructions. > On another note: Anyone tried installing the new 6.2.14 (or something like > that) of BOINC? I installed it without problems on my Linux machines, but [quoted text clipped - 9 lines] > Anyone know the deal on this? What should I pick, and can I move the > Seti@Home WUs across to the new dir? Or should I do something else first? Yeah, when you install 6.2.14 you will lose any un-processed WUs that 5.10.45 had. After the install you will have to re-attach to the SETI project. It will download the stock app and new WUs. So you will have to put your optimized client and app_info.xml in the new directory (...Data\projects\setiathome.berkeley.edu) |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.