Message boards :
Number crunching :
AP V7
Message board moderation
Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 20 · Next
Author | Message |
---|---|
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
OK, AP-credits syndrome si cured. At Beta my 6870 averaged 26min and 612 credit over 192 tasks with AP v7. With AP v6 it was averaging closer to 35min. I'm not at home so I don't have the average credit in front of me, but I think the average was around 640. The gives AP v7 an average of 0.392307 credit a second & AP v6 an average of 0.304761 credit per second. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
David S Send message Joined: 4 Oct 99 Posts: 18352 Credit: 27,761,924 RAC: 12 |
Simple question (I'll get to the complex ones later): I have excluded AP6 from running on my GPU-only cruncher because it uses more of the CPU than I like. Does AP7-GPU use as much of the CPU as 6? David Sitting on my butt while others boldly go, Waiting for a message from a small furry creature from Alpha Centauri. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Simple question (I'll get to the complex ones later): OpenCL AP v7 uses less CPU than AP v6. And the bigger blanking % of task the bigger this difference. Here is illustration of AP v7 behavior: http://lunatics.kwsn.net/1-discussion-forum/astropulse-v7-performance-illustration.0.html For APv6 pictures look that forum for older threads. |
David S Send message Joined: 4 Oct 99 Posts: 18352 Credit: 27,761,924 RAC: 12 |
Simple question (I'll get to the complex ones later): Okay, maybe I'll turn it loose once there's a Lunatics installer for it. David Sitting on my butt while others boldly go, Waiting for a message from a small furry creature from Alpha Centauri. |
OTS Send message Joined: 6 Jan 08 Posts: 369 Credit: 20,533,537 RAC: 0 |
I went to both “Mike’s World†and “Crunchers Anonymous†and did not see any AP7 for Linux 32 bit CPU. Am I just not looking in the right place for those optimized apps? Is there a download site for the source code package that I could use to compile my own binary? |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
I went to both “Mike’s World†and “Crunchers Anonymous†and did not see any AP7 for Linux 32 bit CPU. Am I just not looking in the right place for those optimized apps? Is there a download site for the source code package that I could use to compile my own binary? If there is something caching between the servers and you try adding a ? to the end of the address. http://mikesworldnet.de/download.html vs. http://mikesworldnet.de/download.html?. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
I asked before but I hope it doesn't hurt to ask again: while we wait for the custom builds, does anyone know of the links to the stock Linux applications for AP v7, please? Not a big deal if no-one knows, I suppose. Soli Deo Gloria |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I asked before but I hope it doesn't hurt to ask again: while we wait for the custom builds, does anyone know of the links to the stock Linux applications for AP v7, please? Five minutes and I can find them, my C2D T8100 has had four of the five CPU apps so far. http://boinc2.ssl.berkeley.edu/sah/download_fanout/astropulse_7.00_i686-pc-linux-gnu http://boinc2.ssl.berkeley.edu/sah/download_fanout/astropulse_7.04_i686-pc-linux-gnu__sse http://boinc2.ssl.berkeley.edu/sah/download_fanout/astropulse_7.04_i686-pc-linux-gnu__sse2 http://boinc2.ssl.berkeley.edu/sah/download_fanout/astropulse_7.00_x86_64-pc-linux-gnu http://boinc2.ssl.berkeley.edu/sah/download_fanout/astropulse_7.04_x86_64-pc-linux-gnu__sse2 Claggy |
OTS Send message Joined: 6 Jan 08 Posts: 369 Credit: 20,533,537 RAC: 0 |
I asked before but I hope it doesn't hurt to ask again: while we wait for the custom builds, does anyone know of the links to the stock Linux applications for AP v7, please? Based on the URLs those appear to be the standard apps. Is there a link to the optimized versions for Linux? |
Juha Send message Joined: 7 Mar 04 Posts: 388 Credit: 1,857,738 RAC: 0 |
I asked before but I hope it doesn't hurt to ask again: while we wait for the custom builds, does anyone know of the links to the stock Linux applications for AP v7, please? In addition to Claggy's list, here's Nvidia and Ati apps. Sizes and md5sums not verified from BOINC's downloads. astropulse_7.04_x86_64-pc-linux-gnu__opencl_nvidia_100 , 2487712 bytes , 2858ccde7fe804b73b1478fd05d479c2 astropulse_7.04_x86_64-pc-linux-gnu__opencl_ati_100 , 2483616 bytes , f9ed2869d96d6b6dc89624c434e3534b I'm assuming all the different cuda/nvidia//cc1/opencl versions share the same binary. |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
Based on the URLs those appear to be the standard apps. Is there a link to the optimized versions for Linux? The SSE and SSE2 apps are the Optimised apps. (That's how the Boinc app naming system works) http://setiathome.berkeley.edu/result.php?resultid=3777172054 <core_client_version>7.2.47</core_client_version> The scheduler should send out each compatible app, along with a Wu, once 11 validations are done for each app version, the scheduler should only send out Wu's for the fastest app version (most of the time), If the Stock 7.00 app reaches it's 11 validations first, you may want to abort any work received for that app version to encourage the Optimised 7.04 app versions (each abort lowers the max task per day, when it's low enough the scheduler will send work from another app version) Claggy |
OTS Send message Joined: 6 Jan 08 Posts: 369 Credit: 20,533,537 RAC: 0 |
Based on the URLs those appear to be the standard apps. Is there a link to the optimized versions for Linux? Thank you for the explanation. I thought Berkeley developed apps and then other developers associated Lunatics improved on them and released them as optimized apps on their site, not Berkeley’s. At least that is what I remember when I obtained AK_V8_linux32_ssse3 which I used for many years? I do not recall ever seeing that on Berkeley’s pages but perhaps my memory is failing me. Thanks again and I will give those apps a try. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Now we cooperate with Berkeley's team much more closely. Something still improves over time. |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
In addition to Claggy's list, here's Nvidia and Ati apps. Sizes and md5sums not verified from BOINC's downloads. Thank you Juha and Claggy. Those were what I was looking for. Soli Deo Gloria |
Cliff Harding Send message Joined: 18 Aug 99 Posts: 1432 Credit: 110,967,840 RAC: 67 |
I'm now have 24 AP v7 GPU validated tasks listed. But the 21:42:15 estimated run time still exists for the remaining 103 AP tasks, 4 of which are running all running high priority. This is well beyond the supposed level of 11 tasks required for balancing the estimated run time. With the remaining 99 tasks all having the expected deadline/report date of 11/04, they also will be running in a HP state. Questions - When does the scheduler update the estimated run times for a specific app on a specific host? Is anyone else experiencing this situation? I don't buy computers, I build them!! |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
I'm now have 24 AP v7 GPU validated tasks listed. But the 21:42:15 estimated run time still exists for the remaining 103 AP tasks, 4 of which are running all running high priority. This is well beyond the supposed level of 11 tasks required for balancing the estimated run time. With the remaining 99 tasks all having the expected deadline/report date of 11/04, they also will be running in a HP state. Questions - When does the scheduler update the estimated run times for a specific app on a specific host? Is anyone else experiencing this situation? The estimated time will be for new tasks you download. Tasks currently in your queue retain their estimated run time. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
I'm now have 24 AP v7 GPU validated tasks listed. But the 21:42:15 estimated run time still exists for the remaining 103 AP tasks, 4 of which are running all running high priority. This is well beyond the supposed level of 11 tasks required for balancing the estimated run time. No, you only have 13 non-Outlier tasks completed, not 24, the 'Consecutive valid tasks' value includes all the outlier tasks, tasks that end early don't count. http://setiathome.berkeley.edu/host_app_versions.php?hostid=5501972 Since you're only just past your 11 validated tasks, try another request for GPU work, it should get updated when you receive some. (It should have already done so, since you're received two Wu's since you posted that) Claggy |
Cliff Harding Send message Joined: 18 Aug 99 Posts: 1432 Credit: 110,967,840 RAC: 67 |
I'm now have 24 AP v7 GPU validated tasks listed. But the 21:42:15 estimated run time still exists for the remaining 103 AP tasks, 4 of which are running all running high priority. This is well beyond the supposed level of 11 tasks required for balancing the estimated run time. You lost me with the 13 non-Outlier completed. When looking at my tasks list for this host, I see now 25 validated tasks http://setiathome.berkeley.edu/results.php?hostid=5501972&offset=0&show_names=0&state=4&appid=20. How do I distinguish between non & Outlier tasks? What determines an early task? The only ones that I see that have a short run time are 100% blanked. The scheduler still hasn't updated my manager, as I have not only requested an update manually, but also have stopped and started BOINC with the same results. I don't buy computers, I build them!! |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
I'm now have 24 AP v7 GPU validated tasks listed. But the 21:42:15 estimated run time still exists for the remaining 103 AP tasks, 4 of which are running all running high priority. This is well beyond the supposed level of 11 tasks required for balancing the estimated run time. If you looks at the Application Details for your host you will see the number of tasks completed. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
You lost me with the 13 non-Outlier completed. When looking at my tasks list for this host, I see now 25 validated tasks http://setiathome.berkeley.edu/results.php?hostid=5501972&offset=0&show_names=0&state=4&appid=20. How do I distinguish between non & Outlier tasks? What determines an early task? The only ones that I see that have a short run time are 100% blanked. The scheduler still hasn't updated my manager, as I have not only requested an update manually, but also have stopped and started BOINC with the same results. An outlier is any task that has more Blanking than 5 or 10% (I forget what the limit was), a task with 100% Blanking (it only takes a few seconds, having it included would really inflate the APR), and tasks that exit early like this one: http://setiathome.berkeley.edu/result.php?resultid=3777684842 state.fold_buf_size_short=65536; state.fold_buf_size_long=262144 Found 30 single pulses and 30 repeating pulses, exiting. percent blanked: 0.00 Or this one: http://setiathome.berkeley.edu/result.php?resultid=3777507470 single pulses: 29 repetitive pulses: 30 percent blanked: 2.39 This 2nd Wu reached a it's limit of 30 repetive pulses, then would have sped up, and would have been excluded because of that, the whole point of excluding these Wu's is to get a stable APR. Claggy |
©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.