Message boards :
News :
Nbody WU Flush
Message board moderation
Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 22 · Next
Author | Message |
---|---|
Send message Joined: 22 May 11 Posts: 71 Credit: 5,685,114 RAC: 0 |
Yes. There are no 32bit n-body apps compiled. |
Send message Joined: 8 Nov 11 Posts: 205 Credit: 2,900,464 RAC: 0 |
Yes. There are no 32bit n-body apps compiled. Thanks problem solved. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
I haven't used 32bit anything for a decade, unless no 64bit program is available. Where did you manage to get that from?Yes. There are no 32bit n-body apps compiled. |
Send message Joined: 12 Nov 21 Posts: 236 Credit: 575,038,236 RAC: 0 |
Ryzen 9 computer lists 72 n-body tasks and 1071 separation task as in progress. yep. 72 tasks at 1.25 minutes/task will finish in 57 minutes. I am now down to 49 tasks ready to run. the well has run dry many times. the question still remains from here: https://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=4881&postid=73361 why is the number of n body tasks increasing, while at the same time, telling me there are none available? |
Send message Joined: 22 May 11 Posts: 71 Credit: 5,685,114 RAC: 0 |
Maybe separation tasks block n-body? |
Send message Joined: 12 Nov 21 Posts: 236 Credit: 575,038,236 RAC: 0 |
Pah! I have a Ryzen 9 3900XT. I don't overclock though, had nothing but trouble playing with that in the past. Things crash, get corrupted, all for days of fiddling and a 10% speed increase. If it was meant to go that fast reliably, they'd sell it at that speed. It ran the cinebench r23 stress test for 30 minutes with no issues. There were a number of crashes before it got dialed in, but has been very stable ever since. CPU temps are around 81C at full load, limit is 95. That's with the stock Wraith Prism cooler, and the glass side of the case taken off. Sorry you has all those troubles. there are some good you tube videos on using ryzen master. you might have a go at it again. And of course it will help with your power bill issue, too. NOT! |
Send message Joined: 12 Nov 21 Posts: 236 Credit: 575,038,236 RAC: 0 |
Maybe separation tasks block n-body? nope. when I try to down load n body, separation tasks are unchecked in preferences. |
Send message Joined: 12 Nov 21 Posts: 236 Credit: 575,038,236 RAC: 0 |
Maybe separation tasks block n-body? could there be a hidden limit on the number of n body jobs I can hold in my que? |
Send message Joined: 8 Nov 11 Posts: 205 Credit: 2,900,464 RAC: 0 |
I haven't used 32bit anything for a decade, unless no 64bit program is available. Where did you manage to get that from?Yes. There are no 32bit n-body apps compiled. Have had the machine for many years, I have in fact got the mother board manual for it and I can’t see why it wasn’t 64 bit as it seems to support both. It has run and still runs a lot of Boinc projects as most of them seem to support 32 & 64 bit Silly question can I set up another partition on the same box and generate a 64bit image ? I have lots of disk space available. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
Possibly. I've seen limits where it tells you in the messages tab, and sometimes it's obvious the server is hiding something from me (both with other projects). But with Milkyway, I think you can get enough nbody to run for a few days, even on a good Ryzen.Maybe separation tasks block n-body? |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
Nah, not worth the hassle. If I was to double the speed, maybe I'd try, but not for the small increase you get.Pah! I have a Ryzen 9 3900XT. I don't overclock though, had nothing but trouble playing with that in the past. Things crash, get corrupted, all for days of fiddling and a 10% speed increase. If it was meant to go that fast reliably, they'd sell it at that speed. And of course it will help with your power bill issue, too. NOT!CPUs don't use much power. I have 8 graphics cards. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
So you mean the OS is 32 bit? It it's Windows, I believe it's impossible to make it 64 bit without a clean install. Ridiculous. So I guess you could set it up on a another partition and have both until you've migrated if you ever do.I haven't used 32bit anything for a decade, unless no 64bit program is available. Where did you manage to get that from?Yes. There are no 32bit n-body apps compiled. |
Send message Joined: 8 Nov 11 Posts: 205 Credit: 2,900,464 RAC: 0 |
Thanks I already have in addition a brand new 64 bit I7 powered machine doing work as well, the old one is still functioning so it can contribute albeit slowly. |
Send message Joined: 12 Nov 21 Posts: 236 Credit: 575,038,236 RAC: 0 |
Possibly. I've seen limits where it tells you in the messages tab, and sometimes it's obvious the server is hiding something from me (both with other projects). But with Milkyway, I think you can get enough nbody to run for a few days, even on a good Ryzen. how? what are the recommended settings? 5/7/2022 1:57:37 PM | Milkyway@Home | Starting task de_nbody_08_31_2021_v176_40k__data__13_1647295263_1432084_2 since starting my last n body task, there have been 15 messages saying this: 5/7/2022 1:58:26 PM | Milkyway@Home | Scheduler request completed: got 0 new tasks 5/7/2022 1:58:26 PM | Milkyway@Home | No tasks sent 5/7/2022 1:58:26 PM | Milkyway@Home | No tasks are available for Milkyway@home N-Body Simulation something is really messed up. |
Send message Joined: 11 Mar 22 Posts: 42 Credit: 21,902,543 RAC: 0 |
The current figures indicate, that there is something waaaayy wrong with the MW server setup: Tasks ready to sent : 12033635 N-Body: 12023834 Separation: 9843 I recommend to stop all servers, do a clean installation with config files that are not messed up by "adjustments" and "tweaks" and start all over again. |
Send message Joined: 8 Nov 11 Posts: 205 Credit: 2,900,464 RAC: 0 |
Strangely enough the unsents are starting to drop for Nbody…maybe it had reached a peak ? It’s dropped over 35,000 since you did your message. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
That isn't you. The server is clogged up with stuff to send. It may eventually clear, or it may need Tom to do something. Mind you, I just tested it, using a 10 day buffer with 4 fast cores = 6 minutes per task, so in theory I could get 2400 Nbody tasks. I got 100, then 0 1.5 minutes later, then 0 1.5 minutes after that. Perhaps a 100 task limit has been imposed? That's a very round number I got.Possibly. I've seen limits where it tells you in the messages tab, and sometimes it's obvious the server is hiding something from me (both with other projects). But with Milkyway, I think you can get enough nbody to run for a few days, even on a good Ryzen. Every single one is a _2 task, so presumably we have a lot of finishing off to do as wingmen. So those cannot be cleared or the research would have gaps. Unless Tom can put through a smaller number at a time so as not to confuse the server? |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
The current figures indicate, that there is something waaaayy wrong with the MW server setup:Yeah people tell me to just reinstall Windows too when there's a bug. And then spend two months getting it back to the way it was. Not a good idea. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
Strangely enough the unsents are starting to drop for Nbody…maybe it had reached a peak ?As long as he's stopped it generating new ones, we should be able to slowly work through the checking ones and it will slowly speed up. I do see this: "nbody_work_generator (milkyway_nbody) Running" - although hopefully it won't generate anything when it sees the big queue already there. |
Send message Joined: 16 Mar 10 Posts: 213 Credit: 108,394,388 RAC: 3,530 |
Strangely enough the unsents are starting to drop for Nbody…maybe it had reached a peak ?As long as he's stopped it generating new ones, we should be able to slowly work through the checking ones and it will slowly speed up. In answer to Septimus: It should have reached a peak, as you suggest! And there's a reason for this... The work units that got stalled with a "Didn't need" task would have been marked with a "safety net" time after which the transitioner would look at them again; because of the way that time is calculated, the delay would range from 10 days to 18 days (unless something else (such as a late [post-suspend] reply from the first task) caused the transitioner to look earlier. The "flush" Tom did completed some time on the evening of 19th April - 18 days from then is the evening of 7th May. So the general trend should now be downwards -- at present the unsent count seems to be dropping by about 12,500 to 13,000 an hour. At that rate of decrease, it would take about 3 days, 6 hours to clear one million results - speculative, of course! (And if the rate starts going up rapidly over a prolonged period , time to start really worrying again!...) And yes, as Peter suggests, it'll probably speed up a bit as the queue decreases, though whether that will manifest as more tasks running (perhaps by making things easier for the feeder?) or just as a general improvement in server performance remains to be seen! And, in [further] answer to Peter, I can't find any evidence that there are new N-Body work units being added at present -- admittedly that's by random sampling of high work unit numbers and watching their rate of increase but, unless there's some way of getting "work units generated in the last hour" (or similar) onto the Server Status page, user research can't do any better. Tom did say there's supposed to be a limit, but that doesn't explain how it ran away in the first place, so a little doubt-based research is in order :-) Cheers - Al. |
©2024 Astroinformatics Group