Welcome to MilkyWay@home

Posts by Cliff Harding

1) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 69508)
Posted 2 Feb 2020 by Cliff Harding
Post:
I'm running basically the same setup in SETI, with 4 GPU tasks and 1 CPU task. The manager is set at 70% CPU. I'm also running current Lunatics, and the manager is 7.6.13 (x64), but I don't think that either should be relevant to my situation.

When SETI has no GPU tasks available, I had hoped to run Milkyway the same way.


Your
"(running 0.96 CPUs + 1 NVIDIA device 0)) and device 1."
does not equate with your app config of
"<gpu_usage>.5</gpu_usage>
<cpu_usage>1</cpu_usage>"
It has not taken into account what you told it.
Restart the computer, then you're sure it's read the config you gave it.
Then if it still says 0.96 and 1 instead of 1 and 0.5, then abort all tasks and let it get new ones.
If it still doesn't work, you might have made a slight error in the config file. I've done that before, the slightest mistake in brackets etc and it ignores the whole lot.

I have done the same as you with Milkyway and Einstein as a backup (as Milkyway has 10 minute gaps between batches of tasks it will hand out).


Problem resolved. I keep certain files in a separate folder for use at a later time in case something catastrophic happens. It seems that when I copied this file into my project folder, I didn't pay attention to the full file name. Adjusted the name and it is working as expected -- A man's mind grows dim in his later years.
2) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 69505)
Posted 2 Feb 2020 by Cliff Harding
Post:
It seems that the app_config.xml is being completely ignored. There is no line in the event log that states that the manager recognized it.
3) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 69503)
Posted 2 Feb 2020 by Cliff Harding
Post:
I'm running basically the same setup in SETI, with 4 GPU tasks and 1 CPU task. The manager is set at 70% CPU. I'm also running current Lunatics, and the manager is 7.6.13 (x64), but I don't think that either should be relevant to my situation.

When SETI has no GPU tasks available, I had hoped to run Milkyway the same way.
4) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 69501)
Posted 1 Feb 2020 by Cliff Harding
Post:
I'm attempting to run 2 WUs each on twin NVidia 1070s only, no CPU tasks, on an Intel 7700K under the current Win 10 Pro with the following app_config.xml
<app_config>
<app>
<name>milkyway</name>
<max_concurrent>4</max_concurrent>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>1</cpu_usage>
</gpu_versions>
</app>
</app_config>

I'm keep getting the following line in the event log
02/01/2020 17:49:59 | Milkyway@Home | Your settings do not allow fetching tasks for CPU. To fix this, you can change Project Preferences on the project's web site.

Currently, it's defaulting to (running 0.96 CPUs + 1 NVIDIA device 0)) and device 1.
What to I need to rectify this?
5) Message boards : Number crunching : Its' been a long time! (Message 67933)
Posted 9 Dec 2018 by Cliff Harding
Post:
It's an Intel i7/7700K (4 core / 8 threads) and yes, the machine has been recycled since I started crunching M/W.
6) Message boards : Number crunching : Its' been a long time! (Message 67931)
Posted 8 Dec 2018 by Cliff Harding
Post:
Its' been a long tome since I crunched here and I'm wondering if my GTX 1070 is working at its' max. My app_config.xml is

<app_config>
<app>
<name>milkyway</name>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>1</cpu_usage>
</gpu_versions>
</app>
</app_config>

, but I'm only running 1 task per instead of 2. Any suggestions?

https://milkyway.cs.rpi.edu/milkyway/results.php?hostid=761659&offset=0&show_names=0&state=1&appid=
7) Message boards : Number crunching : Need Help! (Message 61289)
Posted 1 Mar 2014 by Cliff Harding
Post:
Currently I'm running milkyway_separation_1.02_windows_x86_64__opencl_nvidia.exe on my GPUs, what other types/files run on cuda/OpenCl?
8) Message boards : Number crunching : Need Help! (Message 61283)
Posted 1 Mar 2014 by Cliff Harding
Post:
Been away for a while and things have gotten confused when attempting the app_info.xml and I need some help.

02/27/2014 07:46:49 | Milkyway@Home | Syntax error in app_info.xml, but XML Maker 2.2 states there is no syntax error


What program are you using to 'make them'? Since you are using Windows use NOTEPAD ONLY to make the files as word processing progressing programs add hidden stuff that Boinc doesn't like.


The syntax error was caused by the missing tags <app_info> , </app_info>. As far as using XML Maker to create/edit my .XML files, it was suggested by several people at SETI a couple of years ago when I was having problems with syntax.
9) Message boards : Number crunching : Need Help! (Message 61244)
Posted 27 Feb 2014 by Cliff Harding
Post:
Been away for a while and things have gotten confused when attempting the app_info.xml and I need some help.

02/27/2014 07:46:49 | Milkyway@Home | Syntax error in app_info.xml, but XML Maker 2.2 states there is no syntax error

<app>
<name>milkyway</name>
</app>
<file_info>
<name>milkyway</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>102</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>cuda</plan_class>
<coproc>
<type>NVIDIA</type>
<count>0.5</count>
</coproc>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_separation_1.02_windows_x86_64__opencl_nvidia.exe</file_name>
<main_program/>
</file_ref>
</app_version>

[edit] The app_config.xml file that I created seems to be working correctly, running 2 WU's each on 2 x GTX660SC @ 2 Gb each.

<app_config>
<app>
<name>milkyway_separation_modified_fit</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>
</gpu_versions>
</app>
</app_config>
10) Message boards : Number crunching : MW Separation Modified Fit (Message 59453)
Posted 25 Jul 2013 by Cliff Harding
Post:
Hmmm...

Are you using all the CPU cores to run tasks? I found I have to leave one free to drive MW, Collatz, and SAH to max GPU utilization. That gives me 98-99% for MW, and Collatz, and generally better than 95% for SAH (IOW's BOINC restricted to 7 of 8 on the FX-8350).

EAH is a different story. Apparently, it needs far more CPU support for the GPU so I have it spec'ed so when more than 4 want to run it will grab another CPU to help out and even then I can only get to about 85% by running 4 each on my dual 7970's.

<edit> Oh yeah, the MW, Collatz, and SAH data was for running 4 each on the duals, as well.


I'm running BOINC @ 90% (7 cores). Normally I run only S@H AP (both CPU & GPU) on this machine. When the GPU tasks run out, then MW kicks in to fill the GPU gap. Since my earlier post, I've figured out what was needed for the MW app_config.xml file. Right now I running 5 S$H AP CPU tasks, with 2 each MW Separation Modified Fit per GPU, and the GPUs are running at 99%. Whenever I will get some MW OpenCL tasks, they too will run with .5 core reserved for each task and is set up to run 4 tasks per GPU.
11) Message boards : Number crunching : MW Separation Modified Fit (Message 59446)
Posted 24 Jul 2013 by Cliff Harding
Post:
Not sure if this is the right thread, but will try anyway. Started working on the MW Separation Modified Fit (OpenCL only) on 2 x GTX660SC @ 2Gb. Each task is currently using .503 core & 1 GPU. The problem is they are using only approx. 69-75% of the GPU. What do I need to have the CPU & GPU set to .5, so that I reserve a half core CPU, and still run multiple (2) tasks per GPU attempting to get 99% usage?

I am also working Milkyway (OpenCL only) using the default app_info.xml, increasing the GPU to 3 tasks in an attempting to reach 99% usage.

app_config>
<app>
<name>milkyway</name>
<gpu_versions>
<gpu_usage>0.33</gpu_usage>
<cpu_usage>0.5</cpu_usage>
</gpu_versions>
</app>
</app_config>
12) Message boards : Number crunching : Event Log Flags (Message 57344)
Posted 24 Feb 2013 by Cliff Harding
Post:
Since the main program we crunch under is Boinc, at BOTH Seti and here, the debug flags would be identical.


Ok, I'm running 7.0.52 and thought I had a bug. Never really looked at the complete log when Milkyway is running as I mostly run SETI and concentrate all of my log views on that project. It is very disconcerting not to see any down/up messages in the log to do a spot check on which tasks have been transferred. Case in point, on 23 Feb at the beginning of the SETI outage I d/l'ed some Milkyway tasks. Later to cut down on the failed SETI transfer entries in the log I set the network activity to NNT, forgetting about Milkyway. When I noticed that I was not getting any new tasks from here, I clicked on the SHOW ALL TASKS button and saw a load of tasks waiting to report. I allowed network activity to resume, and that is when I noticed that there were no up/down messages. The only way that I can see that files were in fact sucessfully being transferred was to turn on the TASK DEBUG flag.


When I go into the Boinc Manager, down by the clock, and then Advanced, Event Log I see things like this:
2/24/2013 11:12:39 AM | Milkyway@Home | Scheduler request completed: got 1 new tasks
2/24/2013 11:13:44 AM | Milkyway@Home | Sending scheduler request: To fetch work.
2/24/2013 11:13:44 AM | Milkyway@Home | Requesting new tasks for ATI
2/24/2013 11:13:46 AM | Milkyway@Home | Scheduler request completed: got 0 new tasks
2/24/2013 11:13:46 AM | Milkyway@Home | No tasks sent
2/24/2013 11:13:46 AM | Milkyway@Home | This computer has reached a limit on tasks in progress
2/24/2013 11:16:11 AM | Milkyway@Home | Computation for task de_separation_20_sSgr_1_1358941502_14466518_0 finished
2/24/2013 11:16:11 AM | Milkyway@Home | Starting task de_separation_18_sSgr_1_1358941502_14467730_0 using milkyway version 102 (opencl_amd_ati) in slot 1
2/24/2013 11:16:12 AM | Milkyway@Home | Sending scheduler request: To fetch work.
2/24/2013 11:16:12 AM | Milkyway@Home | Reporting 1 completed tasks
2/24/2013 11:16:12 AM | Milkyway@Home | Requesting new tasks for ATI
2/24/2013 11:16:14 AM | Milkyway@Home | Scheduler request completed: got 1 new tasks

Could you just be looking in a different place?



All well and good, I have seen the same thing, but what I don't see are specific lines for the transferred files as in the following examples from Einstein & SETI.

8-Feb-2013 18:46:11 [SETI@home] Started download of ap_02dc12ad_B3_P1_00194_20130218_17530.wu
18-Feb-2013 18:46:11 [SETI@home] Started download of ap_02dc12ad_B3_P0_00232_20130218_16269.wu
18-Feb-2013 18:46:11 [SETI@home] Started download of ap_30dc12ae_B6_P1_00335_20130218_12054.wu

02/24/2013 08:40:07 | Einstein@Home | Started upload of p2030.20121015.G193.75-02.47.C.b0s0g0.00000_80_1_0
02/24/2013 08:40:09 | Einstein@Home | Finished upload of p2030.20121015.G193.75-02.47.C.b0s0g0.00000_80_1_0

13) Message boards : Number crunching : Event Log Flags (Message 57342)
Posted 24 Feb 2013 by Cliff Harding
Post:
Since the main program we crunch under is Boinc, at BOTH Seti and here, the debug flags would be identical.


Ok, I'm running 7.0.52 and thought I had a bug. Never really looked at the complete log when Milkyway is running as I mostly run SETI and concentrate all of my log views on that project. It is very disconcerting not to see any down/up messages in the log to do a spot check on which tasks have been transferred. Case in point, on 23 Feb at the beginning of the SETI outage I d/l'ed some Milkyway tasks. Later to cut down on the failed SETI transfer entries in the log I set the network activity to NNT, forgetting about Milkyway. When I noticed that I was not getting any new tasks from here, I clicked on the SHOW ALL TASKS button and saw a load of tasks waiting to report. I allowed network activity to resume, and that is when I noticed that there were no up/down messages. The only way that I can see that files were in fact sucessfully being transferred was to turn on the TASK DEBUG flag.
14) Message boards : Number crunching : Event Log Flags (Message 57323)
Posted 23 Feb 2013 by Cliff Harding
Post:
Are there any debug/message type flags that are similar to those used in SETI that will appear in the event log, and where would I put them?
15) Message boards : Number crunching : After driver update all gpu wu's fail (Message 57216)
Posted 11 Feb 2013 by Cliff Harding
Post:
I am having a similar problem with nVidia 313.96, but I'm getting valid & invalid.

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=491386&offset=0&show_names=0&state=3&appid=
16) Message boards : News : N_body (Message 56170)
Posted 12 Nov 2012 by Cliff Harding
Post:
Only run Milkyway when running out of work for other projects. After attaching to the project and noticed the following on the N-BODY tasks (System is i7/950, Win7/64-bit, 6Gb ram 1 x EVGA GTX660SC 2Gb, 1 x EVGA GTX460SE 2 1Gb, Nvidia 310.33):

11/12/2012 16:23:57 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task ps_nbody_plus_slice_emd_1_1352203202_7185_0 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:23:57 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task de_nbody_plus_slice_emd_1_1352203202_367_1 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:23:57 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task ps_nbody_plus_slice_emd_1_1352203202_7186_0 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:23:59 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task ps_nbody_plus_slice_emd_1_1352203202_7167_0 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:23:59 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task ps_nbody_plus_slice_emd_1_1352203202_6336_1 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:23:59 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task de_nbody_plus_slice_emd_1_1352203202_175_1 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:24:00 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task de_nbody_plus_slice_emd_1_1352203202_1248_2 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:24:00 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task de_nbody_plus_slice_emd_1_1352203202_6994_1 ( - exit code -1073741515 (0xc0000135))
11/12/2012 16:24:00 | Milkyway@Home | [sched_op] Reason: Unrecoverable error for task de_nbody_plus_slice_emd_1_1352203202_1219_3 ( - exit code -1073741515 (0xc0000135))

Have set project to NNT.
[edit] There was no run time on these tasks. Tasks start and immediately ended. [/edit]
17) Message boards : Number crunching : CPU Scheduling question!! (Message 53140)
Posted 13 Feb 2012 by Cliff Harding
Post:
Are you talking about 1 mw over all cores (multithreaded app like mw nbody) or 1 mw per core (makes 6 * mw)?


Sorry 'bout that. I'm talking 6 tasks, 1 per core. It seems that it only applies to MW 1.00. Nbody is not affected in this manner, which appears to have a normal mix of between cores; meaning Nbody will not overtake more than its resource share.
18) Message boards : Number crunching : CPU Scheduling question!! (Message 53137)
Posted 13 Feb 2012 by Cliff Harding
Post:
Currently running both SETI (100%) and Milkyway (50%) on my i7/950 machine, running on 6 of 8 cores. I have noticed that when Milkyway 1.00 tasks are running they are running on all available cores, placing the SETI tasks in 'waiting to run' state. Is this normal? I was expecting to see a mix of both SETI & Milkyway tasks executing at the same time. Deadline for the Milkyway tasks is far enough away to see this happening.

It should be noted that the machine is running BOINC 7.0.15 and there might be a scheduler problem, but this situation has been in existance even in the 6.10 & 6.12 versions.
19) Message boards : Number crunching : OpenCL video card assignment issue (Message 53115)
Posted 12 Feb 2012 by Cliff Harding
Post:
Thanks again arkayn, your opencl app and app_info are working great so far. I just hope it wasn't a mistake upgrading to Boinc 7.


What version were you using prior to the upgrade? I was using 6.12.34 prior to ugrading to 7.0.8 -> 7.0.15 on my A-SYS without any problems except that one of my devices EVGA GTS250 was not double precision compatible and the WUs were aborting. With the exclude_gpu option in the cc_config.xml file for that device I was able to prevent it for getting opencl work. This option will work for your situation as well.
20) Message boards : News : New server test post (Message 52221)
Posted 5 Jan 2012 by Cliff Harding
Post:
Finally connected to server to report my tasks. I had to ALLOW NEW TASKS, then do an update request for it to work. Hadn't been able to report tasks in the last 24-hrs.

I don't know if the accounts are on the same server, but I cannot update my community prefs. I get a HTTP 500 Internal Server error. http://milkyway.cs.rpi.edu/milkyway/edit_forum_preferences_action.php


Next 20

©2020 Astroinformatics Group