Message boards :
Number crunching :
Don't know where it should go? Stick it here!
Message board moderation
Previous · 1 . . . 100 · 101 · 102 · 103 · 104 · 105 · 106 . . . 147 · Next
Author | Message |
---|---|
tazzduke Send message Joined: 15 Sep 07 Posts: 190 Credit: 28,269,068 RAC: 5 |
OMG. Tomorrow is Friday...... Rofl, I see your sense of humour is still intact. Hope you had a good one :-) Mark |
Speedy Send message Joined: 26 Jun 04 Posts: 1643 Credit: 12,921,799 RAC: 89 |
When people have got enough work in their caches is anybody running more than one task on Windows 10 on a RTX 2070 (non-super) and using the machine as a daily driver? Thanks Tom. Thought I would ask here to see what settings people are using if they are running in multiple units at a time. Seasons greetings to everyone who's been festive |
Tom M Send message Joined: 28 Nov 02 Posts: 5124 Credit: 276,046,078 RAC: 462 |
If you look in your Windows Data Folder (hidden) you will find documentation that includes examples for high-end Nvidia's. If you start a separate thread on the topic there are a number of people running Windows with high-end Nvidia's that will probably chime in. Failing all that here is an ancient "best bang for the buck" thread which may offer some useful command line information: https://setiathome.berkeley.edu/forum_thread.php?id=81516#1870570 HTH, Tom A proud member of the OFA (Old Farts Association). |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
When people have got enough work in their caches is anybody running more than one task on Windows 10 on a RTX 2070 (non-super) and using the machine as a daily driver? Now here's a subject I haven't thought about in a while. If you plan to run more than 1 work unit on your GPU, be aware of how much RAM the GPU has. By default, Nvidia only uses about 27% of the available RAM for Scientific Calculations. ATI and Intel is closer to 50-67%. Depending on your command line, you will find that you can quickly eat up that available RAM. For OpenCL MB the best ratio is 3 work units per GPU. For APs, it was 5 APs per GPU. But since its nearly impossible to get that many APs anymore, you can just run the ratio of 3 to 1 to keep them moving in the FIFO instead of waiting till you run out of MBs. Don't over commit the CPU. I also suggest leaving 1-2 thread free to help fee the GPU and leaving the OS something to use as well. I don't recommend the "use at most" % CPU. Instead I use an <project_max_concurrent> in my app_config.xml Yes, Yes, Yes....I know this is going to start a fight over which is the best way to do this. I'm only giving you my opinion. Do what you want. Everyone HAPPY?? All this are on high end GPUs. Those with the fastest RAM on the GPU and the kind of RAM. Go back and look at the DDR used on the 1080Ti vs 1080, 1070 etc. It wasn't the same. This is why we told people to only do these with the highest end GPUs. All this assumes you aren't using it as a daily driver. If you are, then you will find that there may be significant lag to freezing when you use your computer. Use at your own risk. |
iriemon Send message Joined: 22 Aug 16 Posts: 2 Credit: 67,228 RAC: 0 |
Just recently rejoined the SETI effort and just wondering if there is any way to sort the team stats by total credits, rather than average credits? Thanks |
Mr. Kevvy Send message Joined: 15 May 99 Posts: 3776 Credit: 1,114,826,392 RAC: 3,319 |
|
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
Don't know if anyone's come across this, but my Windows machine running 6.10.58 apparently hasn't been able to connect in a while because of a SSL failure. I had dealt with this a few months ago when dealing with that whole openssl debacle where it needed to be updated to at least 0.9.8o, but now ca-bundle.crt was a culprit. I turned all kinds of logging debug flags on and found SSL failures and "unknown cipher" and stuff like that, so.. I went and grabbed the latest .crt from the libcurl download (curl.haxx.se) and dropped it in.. works fine now. But I wonder if this will become a problem again in short order, or if the valid dates can change without the hash key in the .crt having to be updated? 2019-12-29 02:26:46 [http_debug] [ID#1] Info: SSLv3, TLS handshake, Finished (20): 2019-12-29 02:26:46 [http_debug] [ID#1] Info: SSL connection using ECDHE-RSA-AES256-GCM-SHA384 2019-12-29 02:26:46 [http_debug] [ID#1] Info: Server certificate: 2019-12-29 02:26:46 [http_debug] [ID#1] Info: subject: CN=setiathome.berkeley.edu 2019-12-29 02:26:46 [http_debug] [ID#1] Info: start date: 2019-11-19 18:06:32 GMT 2019-12-29 02:26:46 [http_debug] [ID#1] Info: expire date: 2020-02-17 18:06:32 GMT 2019-12-29 02:26:46 [http_debug] [ID#1] Info: subjectAltName: setiathome.berkeley.edu matched 2019-12-29 02:26:46 [http_debug] [ID#1] Info: issuer: C=US; O=Let's Encrypt; CN=Let's Encrypt Authority X3 2019-12-29 02:26:46 [http_debug] [ID#1] Info: SSL certificate verify ok. That's only a 3-month certificate. Usually these things go for 1-10 years. What do you guys think? edit: the one good thing about updating ca-bundle.crt is that I no longer get a successful scheduler work fetch only to have the log immediately show that server communication failed and to check my internet connection. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
iriemon Send message Joined: 22 Aug 16 Posts: 2 Credit: 67,228 RAC: 0 |
Here you go. The "Total Credit" column header should be clickable and provide this. I can't get it to sort. The header isn't clickable... Here is the page URL https://setiathome.berkeley.edu/team_members.php?teamid=30195&offset=0&sort_by=expavg_credit |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Here you go. The "Total Credit" column header should be clickable and provide this. Just change the sort_by in the URL. https://setiathome.berkeley.edu/team_members.php?teamid=30195&offset=0&sort_by=total_credit Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
Greetings, I've been seeing a strange happenstance since the servers were reverted so we can still do anonymous platform. During the day WUs that are ready to report get stacked up where the ones at night do not stack much at all. Just now , about 10-15 minutes ago,I saw that I had over 50 WUs to report. Any ideas? Have a great day and Happy New Year! :) Siran CAPT Siran d'Vel'nahr - L L & P _\\// Winders 11 OS? "What a piece of junk!" - L. Skywalker "Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath |
Wiggo Send message Joined: 24 Jan 00 Posts: 34883 Credit: 261,360,520 RAC: 489 |
You may have had stuck uploads/downloads or you just couldn't connect to the server for some period of time that put you into a backoff state for a while is all I can think of. Cheers. |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1853 Credit: 268,616,081 RAC: 1,349 |
Greetings, Check your BOINC Manager event log. It will tell you if tasks are not being reported and what it thinks the problem is ... |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
You may have had stuck uploads/downloads or you just couldn't connect to the server for some period of time that put you into a backoff state for a while is all I can think of. Hi Wiggo, The thing is, since they fixed the major anonymous platform issue, this has been happening every day since. It's not a random "issue". During the day they stack, over night they don't nearly what they do daily. At night seldom more than 2 or 3 stacked. Daily, I've seen as many as 30+. This is only on my main Linux/Winders PC. No changes made except during the anonymous issue when I went to stock and then back to anonymous. I'm not really that worried about it. I don't see any problems in the Event Log when I check it. Basically I was wondering if this was a server side issue, perhaps. ;) Thanks and have a great day! :) Siran CAPT Siran d'Vel'nahr - L L & P _\\// Winders 11 OS? "What a piece of junk!" - L. Skywalker "Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13746 Credit: 208,696,464 RAC: 304 |
The thing is, since they fixed the major anonymous platform issue, this has been happening every day since. It's not a random "issue".Stuck downloads & uploads taking a couple of attempts have been occurring for a month or 2 now, at least. It's not a new issue, and is no worse now than it was before the Scheduler version problems. Grant Darwin NT |
rob smith Send message Joined: 7 Mar 03 Posts: 22223 Credit: 416,307,556 RAC: 380 |
Just now it is almost certainly a server-side issue. There are several thing placing extra loads on the servers, most obvious being the increased task/processor allocation, but also auto-resend being turned on. It is quite spectacular the additional load, in terms of queries/second that doing automatic resends (or ghost clearance) creates - far more than the routine "failed so to another user" process Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
The thing is, since they fixed the major anonymous platform issue, this has been happening every day since. It's not a random "issue".Stuck downloads & uploads taking a couple of attempts have been occurring for a month or 2 now, at least. It's not a new issue, and is no worse now than it was before the Scheduler version problems. Hi Grant, I haven't been noticing any stuck downloads or uploads except just after a maintenance cycle. I just looked and I have 22 GPU and 4 CPU WUs stacked up waiting to report. Ok, just now they reported not even a minute after I last checked. Weird. Here's a thought: How much control, over BOINC, does BOINCTasks have? BT is running on the PC in question. Have a great day! :) Siran CAPT Siran d'Vel'nahr - L L & P _\\// Winders 11 OS? "What a piece of junk!" - L. Skywalker "Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13746 Credit: 208,696,464 RAC: 304 |
For those with more heat than they can handle. Full immersion cooling for the consumer (unfortunately at an eye watering price). There are two types of immersive systems: two phase, where the liquid turns to a gas on heating and then condenses back into a liquid, or single phase systems that rely on a pump and a radiator to help move the liquid. This is the latter, and if you’ve ever heard of the ‘mineral oil’ PC, this is something very similar, except this is a case and coolant dedicated for immersion systems. Grant Darwin NT |
Tom M Send message Joined: 28 Nov 02 Posts: 5124 Credit: 276,046,078 RAC: 462 |
For those with more heat than they can handle. Wow. A proud member of the OFA (Old Farts Association). |
tazzduke Send message Joined: 15 Sep 07 Posts: 190 Credit: 28,269,068 RAC: 5 |
Greetings All Just throwing it out there, so for my understanding. Could it be that the server backlog on validation and also since the quota's were increased, that this culminated in everybody's RAC increasing at a slower rate than before. Regards Mark |
rob smith Send message Joined: 7 Mar 03 Posts: 22223 Credit: 416,307,556 RAC: 380 |
The long queue for validation will certainly affect everyone's RAC. For most people the number of tasks in-hand will have very little impact as they are working pretty much on one-return = one-supplied now. But there have been a quite a number of major server issues which have slowed/stalled the actual rate of validation and that's what hits our RACs. There may of course be some peole who hear that the servers are "down" and decide to turn off their hosts until the servers are up, and so aren't processing or returning work. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.