Message boards :
Number crunching :
Setting up Linux to crunch CUDA90 and above for Windows users
Message board moderation
Previous · 1 . . . 134 · 135 · 136 · 137 · 138 · 139 · 140 . . . 162 · Next
Author | Message |
---|---|
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 |
Or for those running GPU's that don't support the Special Sauce app. . . Nope, hosts with cards that do not support the special app are far, far better off with SoG than with Cuda60 ... :( Stephen :( |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
Ubuntu's Archive Manager doesn't natively understand 7zip compression. Add it to the Archive Manager with: I’ve never had to install p7zip on Ubuntu 18. Just right click -> extract from the GUI. Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
Sleepy Send message Joined: 21 May 99 Posts: 219 Credit: 98,947,784 RAC: 28,360 |
. . Nope, hosts with cards that do not support the special app are far, far better off with SoG than with Cuda60 ... :(My second card is a GTX660TI, therefore not suitable for CUDA90 (have to check for the other special source, actually, but I do not know if I have enough memory). In any case, to start it off, I ran stock. Survived some struggle with CUDA application sent by the servers, then after a short while the cycle brings you some WU to be processed by SOG. At that point I had everything on board and I applied it to app_info.xml and now no more glitches, it is SOG all the time. I could have done it from the start (and I did, actually), but either I downloaded the wrong package from Lunatics or there was some other problem. Therefore, since I know they are the same, I relied on the server. And it is working well, considering the limits of this old card. Cheers! |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Ubuntu's Archive Manager doesn't natively understand 7zip compression. Add it to the Archive Manager with: I couldn't unpack the AIO when I first got it. Had to install p7zip. But that was a long time ago. Maybe back in 2017 before the release of Ubuntu 18. Could have been on Ubuntu 16. I don't remember. Thought others had issues too unpacking and was resolved by installing p7zip. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
. . Nope, hosts with cards that do not support the special app are far, far better off with SoG than with Cuda60 ... :(My second card is a GTX660TI, therefore not suitable for CUDA90 (have to check for the other special source, actually, but I do not know if I have enough memory). Sorry, your GTX 660 Ti is not suitable for the special app, not even the zi3v CC3.5 one. Your card maxes out at CC3.0 https://developer.nvidia.com/cuda-gpus Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Wiggo Send message Joined: 24 Jan 00 Posts: 35214 Credit: 261,360,520 RAC: 489 |
Now I'm hoping that changing my app_info.xml from its current, <app_info> <app> <name>setiathome_v8</name> </app> <file_info> <name>setiathome_x41p_V0.98b1_x86_64-pc-linux-gnu_cuda90</name> <executable/> </file_info> <app_version> <app_name>setiathome_v8</app_name> <platform>x86_64-pc-linux-gnu</platform> <version_num>801</version_num> <plan_class>cuda90</plan_class> <cmdline></cmdline> <coproc> <type>NVIDIA</type> <count>1</count> </coproc> <avg_ncpus>0.1</avg_ncpus> <max_ncpus>0.1</max_ncpus> <file_ref> <file_name>setiathome_x41p_V0.98b1_x86_64-pc-linux-gnu_cuda90</file_name> <main_program/> </file_ref> </app_version> <app> <name>astropulse_v7</name> </app> <file_info> <name>astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100</name> <executable/> </file_info> <file_info> <name>AstroPulse_Kernels_r2751.cl</name> </file_info> <file_info> <name>ap_cmdline_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100.txt</name> </file_info> <app_version> <app_name>astropulse_v7</app_name> <platform>x86_64-pc-linux-gnu</platform> <version_num>708</version_num> <plan_class>opencl_nvidia_100</plan_class> <coproc> <type>NVIDIA</type> <count>1</count> </coproc> <avg_ncpus>0.1</avg_ncpus> <max_ncpus>0.1</max_ncpus> <file_ref> <file_name>astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100</file_name> <main_program/> </file_ref> <file_ref> <file_name>AstroPulse_Kernels_r2751.cl</file_name> </file_ref> <file_ref> <file_name>ap_cmdline_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100.txt</file_name> <open_name>ap_cmdline.txt</open_name> </file_ref> </app_version> <app> <name>setiathome_v8</name> </app> <file_info> <name>MBv8_8.22r3711_sse41_intel_x86_64-pc-linux-gnu</name> <executable/> </file_info> <app_version> <app_name>setiathome_v8</app_name> <platform>x86_64-pc-linux-gnu</platform> <version_num>800</version_num> <file_ref> <file_name>MBv8_8.22r3711_sse41_intel_x86_64-pc-linux-gnu</file_name> <main_program/> </file_ref> </app_version> <app> <name>astropulse_v7</name> </app> <file_info> <name>ap_7.05r2728_sse3_linux64</name> <executable/> </file_info> <app_version> <app_name>astropulse_v7</app_name> <version_num>704</version_num> <platform>x86_64-pc-linux-gnu</platform> <plan_class></plan_class> <file_ref> <file_name>ap_7.05r2728_sse3_linux64</file_name> <main_program/> </file_ref> </app_version> </app_info> To this, <app_info> <app> <name>setiathome_v8</name> </app> <file_info> <name>setiathome_x41p_V0.98b1_x86_64-pc-linux-gnu_cuda90</name> <executable/> </file_info> <app_version> <app_name>setiathome_v8</app_name> <platform>x86_64-pc-linux-gnu</platform> <version_num>801</version_num> <plan_class>cuda90</plan_class> <cmdline></cmdline> <coproc> <type>NVIDIA</type> <count>1</count> </coproc> <avg_ncpus>0.1</avg_ncpus> <max_ncpus>0.1</max_ncpus> <file_ref> <file_name>setiathome_x41p_V0.98b1_x86_64-pc-linux-gnu_cuda90</file_name> <main_program/> </file_ref> </app_version> <app> <name>setiathome_v8</name> </app> <file_info> <name>MBv8_8.22r3711_sse41_intel_x86_64-pc-linux-gnu</name> <executable/> </file_info> <app_version> <app_name>setiathome_v8</app_name> <platform>x86_64-pc-linux-gnu</platform> <version_num>800</version_num> <file_ref> <file_name>MBv8_8.22r3711_sse41_intel_x86_64-pc-linux-gnu</file_name> <main_program/> </file_ref> </app_version> That I won't get any more AP's or is there something else that I need to change in TBar's AIO as well? (I looked in all the obvious places, but I may have missed something.) They are causing a bit of havoc with a little experiment I'm trying to run ATM and besides, there's plenty of Windows rigs out there that can actually profit from them instead. ;-) Cheers. |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
You can just uncheck the AstroPulse check box on the seti preferences on the website. Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
Tom M Send message Joined: 28 Nov 02 Posts: 5124 Credit: 276,046,078 RAC: 462 |
Seems reasonable based on my limited knowledge of how the app_info.xml file interacts with the Seti Scheduler. Besides if you don't want AstroPulse tasks, that leaves more for ME (yes I also have a Windows box...) :) Tom A proud member of the OFA (Old Farts Association). |
Wiggo Send message Joined: 24 Jan 00 Posts: 35214 Credit: 261,360,520 RAC: 489 |
You can just uncheck the AstroPulse check box on the seti preferences on the website.We've all heard about how well that works these days Ian so I'd just prefer to nip that in the bud from my end. ;-) Cheers. |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
What do you mean? If you uncheck the astropulse box, and you have the computer set to the same location as the location you made the setting change on, the server won’t send you any AP work. Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 |
What do you mean? If you uncheck the astropulse box, and you have the computer set to the same location as the location you made the setting change on, the server won’t send you any AP work. . . That was what I did :). I edited the location settings in SETI preferences for the Linux boxes and unticked the 'Astropulse' option. But because I have left "if there is no work send me anything" option I still get the occasional AP, though I don't understand just why when there is ample work of other types ... I am surprised that wasn't Wiggo's first approach ... Stephen :) |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Everything looks ok for chopping out the AP application. But your app_info.xml is malformed and will dump work because you didn't close your last xml statement. You opened with: <app_info> but didn't close with: </app_info> You can check for malformed XML scripts at: https://www.xmlvalidation.com/index.php Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Wiggo Send message Joined: 24 Jan 00 Posts: 35214 Credit: 261,360,520 RAC: 489 |
Thanks for both Keith, I'll implement that after I get over Good Friday here (I'm surprise that I missed that closing tag, probably because I was trying to get things done here in a hurry before LLB arrived) and I've bookmarked that link. ;-) That'll make sure that my internet gets a break and I don't wind up with that funny behaviour that's been reported by many here when using the internet options. Cheers. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13771 Credit: 208,696,464 RAC: 304 |
What do you mean? If you uncheck the astropulse box, and you have the computer set to the same location as the location you made the setting change on, the server won’t send you any AP work.And will you have random periods where you will struggle to get any MB work as well, that was the only reason I enabled AP work on my systems. To keep getting MB work. Grant Darwin NT |
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 |
And will you have random periods where you will struggle to get any MB work as well, that was the only reason I enabled AP work on my systems. To keep getting MB work. . . I have disabled AP in the location options for my Linux rigs and have no trouble getting new work (mostly, there are still some issues with http errors after each outage). But I do have "when no work of types selected available send me anything" ticked, which is why I occasionally get an AP or 2, but very, very rarely. Stephen |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13771 Credit: 208,696,464 RAC: 304 |
I have disabled AP in the location options for my Linux rigs and have no trouble getting new work (mostly, there are still some issues with http errors after each outage). But I do have "when no work of types selected available send me anything" ticked, which is why I occasionally get an AP or 2, but very, very rarely.Maybe whatever the issue was has been resolved/resolved itself. If so, it's only taken a few years. Grant Darwin NT |
Stephen "Heretic" Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 |
Maybe whatever the issue was has been resolved/resolved itself. If so, it's only taken a few years. . . In the world of astronomy nothing happens fast ... :) I have been waiting for 4 years to crunch data from Parkes ... :( Stephen <shrug>3</shrug> |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
Semi-wierdness. Currently have 7 x GTX980s running on the Big Boy. Went to add #8, and the box won't even POST. Most strange ... typical HP, I guess . . . |
Tom M Send message Joined: 28 Nov 02 Posts: 5124 Credit: 276,046,078 RAC: 462 |
Semi-wierdness. While I could boot more than 7 gpus on a server MB I had, it wasn't stable. 7 was the maximum I could run reliably. So if you want to increase production you may have to resort to another option..... replace Gpu's with later generations.... or change MB/cpu's. The only way I got past 7 gpus was with a 6 slot Biostar Mining MB and an pci-extender. And it wasn't completely stable. I THINK I could have made it stable if I had been willing to run the cpu slower. If you want to go past 7 gpus of "whatever" model you pretty much are going to have to buy a purpose built MB. For instance, Tbar found a 12 gpu slot MB for $30 (Biostar TB250-BTC Pro) and there is always the "Asus B250 Mining Expert" which is widely available has has at least 19 gpu slots with multiple PSU segmentation etc. I found myself reluctant to switch to a LGA 1151 cpu because the ones I could afford were 4c/8t and would basically not be able to crunch other projects on the cpu as well as support the gpus. But Tbar has shown they can and do work well. My understanding is that this rig https://setiathome.berkeley.edu/show_host_detail.php?hostid=6813106 which is crunching at 770,000+ RAC and currently ranked 3rd is an Asus B250 Mining Expert MB. He has 14 gpus running off it right now. 1 RTX 2070, 6 gtx 1070's, 7 gtx 1060 3Gb's. This combo is (probably) the low cost leader in terms of capital costs and total production. There is a thread devoted to this topic: https://setiathome.berkeley.edu/forum_thread.php?id=83307 You could try a lower cost AMD solution (Biostar TA320-BTC) with a really low-cost black friday Amd 2700x ($159) [with cpu cooler] or you could go for a LGA 1151 Biostar TB250-BTC Pro ($64 on NewEgg). Make sure you get the "right" LGA 1151 cpu. Intel has at least 2 different generations of LGA 1151 cpus that don't boot in the "wrong" LGA 1151 socket. HTH, Tom ps. Sometimes updating the Bios of the MB can help. A proud member of the OFA (Old Farts Association). |
Jimbocous Send message Joined: 1 Apr 13 Posts: 1855 Credit: 268,616,081 RAC: 1,349 |
So close, and yet so far ... The Big Boy just hit a RAC of 299,999.90 (seriously!) before falling back. Doubt I'll get it to bust the hoped-for 300k. As George Harrison sings "Isn't it a pity" in my feeble brain :) Meanwhile, its understudy has now proven to me it cannot support more than 4 GPUs and remain stable. 5 and 6 was just a bridge too far, resulting in many "finish file present too long" and 'exceeded time limit" errors. [sorry for the OT-ness] |
©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.