Message boards :
Number crunching :
GBT MESSIER031 work on GPU NVIDIA
Message board moderation
Author | Message |
---|---|
Gianfranco Lizzio Send message Joined: 5 May 99 Posts: 39 Credit: 28,049,113 RAC: 87 |
Good news the Berkeley servers started the distribution of guppi work unit on CUDA NVIDIA GPU. I don't want to believe, I want to know! |
TimeLord04 Send message Joined: 9 Mar 06 Posts: 21140 Credit: 33,933,039 RAC: 23 |
I got some of these into my cache last night on both systems. :-) TimeLord04 Have TARDIS, will travel... Come along K-9! Join Calm Chaos |
Gianfranco Lizzio Send message Joined: 5 May 99 Posts: 39 Credit: 28,049,113 RAC: 87 |
The interesting thing of the overflows is that the result are not simple spikes but triplets. I don't want to believe, I want to know! |
Jeff Buck Send message Joined: 11 Feb 00 Posts: 1441 Credit: 148,764,870 RAC: 0 |
Was just checking my machines before I went to bed and watched one blow through 7 consecutive MESSIER031 tasks in 41 seconds. 4/23/2016 10:06:22 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0008.32263.889.23.46.111_0 4/23/2016 10:06:22 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0008.32263.889.23.46.111_0 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:25 PM | SETI@home | Started upload of 05jn10aa.11353.68280.6.33.31_0_0 4/23/2016 10:06:27 PM | SETI@home | Finished upload of 05jn10aa.11353.68280.6.33.31_0_0 4/23/2016 10:06:27 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0008.32263.889.23.46.111_0 finished 4/23/2016 10:06:27 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0009.32321.835.24.47.181_1 4/23/2016 10:06:27 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0009.32321.835.24.47.181_1 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:29 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0008.32263.889.23.46.111_0_0 4/23/2016 10:06:31 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0008.32263.889.23.46.111_0_0 4/23/2016 10:06:32 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0009.32321.835.24.47.181_1 finished 4/23/2016 10:06:32 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.143_1 4/23/2016 10:06:32 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.143_1 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:34 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0009.32321.835.24.47.181_1_0 4/23/2016 10:06:36 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0009.32321.835.24.47.181_1_0 4/23/2016 10:06:39 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.143_1 finished 4/23/2016 10:06:39 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.173_1 4/23/2016 10:06:39 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.173_1 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:41 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.143_1_0 4/23/2016 10:06:44 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.143_1_0 4/23/2016 10:06:44 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.173_1 finished 4/23/2016 10:06:44 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.195_1 4/23/2016 10:06:44 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.195_1 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:46 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.173_1_0 4/23/2016 10:06:49 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.173_1_0 4/23/2016 10:06:51 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.195_1 finished 4/23/2016 10:06:51 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.201_0 4/23/2016 10:06:51 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.201_0 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:53 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.195_1_0 4/23/2016 10:06:56 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.195_1_0 4/23/2016 10:06:57 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.201_0 finished 4/23/2016 10:06:57 PM | SETI@home | Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.207_0 4/23/2016 10:06:57 PM | SETI@home | [cpu_sched] Starting task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.207_0 using setiathome_v8 version 800 (cuda50) in slot 6 4/23/2016 10:06:59 PM | SETI@home | Started upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.201_0_0 4/23/2016 10:07:01 PM | SETI@home | Finished upload of blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.201_0_0 4/23/2016 10:07:03 PM | SETI@home | Computation for task blc0_2bit_guppi_57397_MESSIER031_0011.1278.840.23.46.207_0 finished 4/23/2016 10:07:03 PM | SETI@home | Starting task 05jn10aa.6973.2112.5.32.212_2 |
BetelgeuseFive Send message Joined: 6 Jul 99 Posts: 158 Credit: 17,117,787 RAC: 19 |
The interesting thing of the overflows is that the result are not simple spikes but triplets. I have a relatively high number of inconclusives on these tasks. All my inconclusive GPU tasks overflow with 30 triplets, but the wingmen for these tasks (CPU) report a high number of triplets (typically 25 or 26) and also a number of spikes. Anyone else notice this ? Tom |
Jeanette Send message Joined: 25 Apr 15 Posts: 55 Credit: 7,827,469 RAC: 0 |
I experience the exact same, though most of mine are still pending. It's the same on both Lunatics and stock GPU tasks. |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
The interesting thing of the overflows is that the result are not simple spikes but triplets. The same here and I'm now wondering what sort of mess these are going cause. Cheers. |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
The interesting thing of the overflows is that the result are not simple spikes but triplets. Most probably it's the result of current validator weakness in overflowed tasks comparison. It relates to only CPU-like sequential style of task processing. But overflows report only part of all signals inside data, so, depend from processing order. Seems CUDA build affected too (and I expect even greater effect on SoG builds). With Arecibo data where overflows were early ones and mostly on spikes there were attempts to simulate serial order of execution on early processing stages, where overflow most probably. This apparently doesn't help a lot with new types of overflows we seeing with GBT data. Until validator will account for different processing order possibility such inconclusives (and some of them will even turn into invalids later though relatively small part) are unavoidable. |
bluestar Send message Joined: 5 Sep 12 Posts: 7016 Credit: 2,084,789 RAC: 3 |
Except for the possible overflows, are triplets and their respective scores perhaps a better indicator than a similar gaussian score? |
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 |
Except for the possible overflows, are triplets and their respective scores perhaps a better indicator than a similar gaussian score? Indicator of what? |
bluestar Send message Joined: 5 Sep 12 Posts: 7016 Credit: 2,084,789 RAC: 3 |
Sorry about that, Raistmer. But perhaps I should have said "intelligent signal" instead, rather than mentioning either triplets or gaussians. Please have me excused, but I was rapped for messing up with the discussion about possible driver errors and overflows being associated with such tasks. Therefore I am not doing such type of business anymore. Things are definitely working here and if something should ever come up, I rather will believe it to be true. You know, Seti@home is among other things dealing with the numbers related to the processing of data coming from space. We divide these data into signal types when next being processed in lakc of somthing better or maybe else. If a triplet or a gaussian is of no scientific value, who are then supposed to be doing the science? The living world could be filled with conspiracies and lies and therefore you should rather believe in science fiction rather than the face of the man in the Moon. As far as I know, the possible detection of an intelligent signal coming from space is by means of the data that are being processed by the Seti@home client and the scores which are being obtained for the different signal types. You may well be a believer of extraterrestrial intelligence, but the subject of UFO's or aliens is not supposed to be the same as the processing of data we are currently are doing, at least not all the time. Make it space or time travel if you will, or even rocket or jet engines for that matter. If you still wish to detect an intelligent signal from the errors the client may be able to produce at times, you are welcome, but doing such a thing does not necessarily give a better explanation for the world or even Universe we happen to be a part of. This is why I am still here. |
©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.