Message boards :
Number crunching :
Astropulse time estimation
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
Batter Up Send message Joined: 5 May 99 Posts: 1946 Credit: 24,860,347 RAC: 0 |
I get 30% more cobblestones for AP GPU WU for the same power consumption as opposed to MB. The worse thing about this is not a few abort MB WU when AP WU are back. |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
You and eveyone else, look this post and follow the thread: http://setiathome.berkeley.edu/forum_thread.php?id=73792&postid=1463003 |
Batter Up Send message Joined: 5 May 99 Posts: 1946 Credit: 24,860,347 RAC: 0 |
You and eveyone else, look this post and follow the thread: It it not a GPU vs. CPU situation but a AP vs. MB WU situation. One does not have to accept any CPU work. If I could get all AP the CPU, 100%, would go to helping the GPUs. |
Michael W.F. Miles Send message Joined: 24 Mar 07 Posts: 268 Credit: 34,410,870 RAC: 0 |
You can quickly fix it by adding a FLOPS entry into the app_info file. You must be careful though, too high a value will result in too short an estimated runtime. This will cause any work in progress to exit with a time exceeded error. An approximate value would be around 40000000000, you would enter it as such; <app_version> <app_name>astropulse_v6</app_name> <version_num>602</version_num> <flops>40000000000</flops> <file_ref> <file_name>ap_6.02r561_OSX64</file_name> <main_program/> It would probably be a good idea to remove one of the zeros on the first attempt if you have any work in progress. You might want to fine tune the number based on what is displayed in the Remaining (estimated) column after restart. A higher number will give a lower estimated time. I got tried of waiting for the system to level this out so I put in 4 billion on the FLOPS estimate and all tasks that were still in my cache went up to 165 hours from 161. All new arrivals are being estimated at 11 hours as it should be The above example is showing 40 billion and my floating point speed on my 1100T Boinc tells me my floating point speed is estimated at 3054.12 million ops/sec so I put in 4000000000 I hope my cache fills up before they shut down the splitters for AP for the week Michael Miles |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
OP, 600 cobblestones credit for 2 seconds work; I would stick with it. Yes a perfect random number generator... |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Says 45.34 GFLOPS here, http://setiathome.berkeley.edu/host_app_versions.php?hostid=6339101 The other day when I made that first post I believe it was around 44 and I gave a setting for 40. Then mentioned Fine Tuning.... It's too late to change it now if you have some tasks estimated at 11 hours and some at 165 hours. |
Michael W.F. Miles Send message Joined: 24 Mar 07 Posts: 268 Credit: 34,410,870 RAC: 0 |
Says 45.34 GFLOPS here, http://setiathome.berkeley.edu/host_app_versions.php?hostid=6339101 The other day when I made that first post I believe it was around 44. I know my APR is 45 gflops but my bench test for this cpu is 3054.12 million ops/sec So that is 4 billion. It's very confusing that is for sure. I put the extra 0 in to make it 45 billion so we what the estimates show At 4 billion the estimate was dead on at 11 hours which is the time it takes in reality. I will keep on it. Question: After I get enough valid tasks in do I then remove the flops entry? Thanks very much TBar Michael Miles |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
If you have some tasks with large differences in estimated times I would leave it alone for now. After all the tasks have close to the same estimated times you can remove it if you wish and see what happens. |
Michael W.F. Miles Send message Joined: 24 Mar 07 Posts: 268 Credit: 34,410,870 RAC: 0 |
TBar Will do thanks bud for all your help Michael Miles |
Michael W.F. Miles Send message Joined: 24 Mar 07 Posts: 268 Credit: 34,410,870 RAC: 0 |
So far cache is full. 96 601 units in the cache. Thanks for the help TBar and all who replied Michael Miles |
©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.