Posts by Claggy

1) Message boards : Cruncher's Corner : Parallella, Raspberry Pi, FPGA & All That Stuff (Message 132936)
Posted 7 days ago by Claggy
Since this device is available now, I would like to ask: has anyone managed to teach this device to crunch (including GPU) ?

Ivan has compiled the Seti MBv7 Cuda app for it.

Claggy
2) Message boards : Problems and Bug Reports : Progress Stucks in 16.666% (Message 132642)
Posted 21 days ago by Claggy
Not all Einstein apps display frequent progress, or checkpoint frequently,

Wait, and you'll see it jump to the next checkpoint/% done.

Claggy
3) Message boards : Cruncher's Corner : Only 2 threads running instead of 8 (Message 132546)
Posted 25 days ago by Claggy
2014-07-28 13:06:18.1280 [PID=2086 ] [version] Best version of app einstein_S6CasA is ID 567 (6.27 GFLOPS)
2014-07-28 13:06:18.1287 [PID=2086 ] [send] stopping work search - insufficient disk space
2014-07-28 13:06:18.1288 [PID=2086 ] [send] stopping work search - insufficient disk space
2014-07-28 13:06:18.1288 [PID=2086 ] [send] stopping work search - insufficient disk space
2014-07-28 13:06:18.1288 [PID=2086 ] [send] stopping work search - insufficient disk space

2014-07-28 13:06:18.1477 [PID=2086 ] [debug] [HOST#10689021] MSG(high) Gravitational Wave S6 Directed Search (CasA) needs 8.86MB more disk space. You currently have 86.51 MB available and it needs 95.37 MB.
2014-07-28 13:06:18.1477 [PID=2086 ] Sending reply to [HOST#10689021]: 0 results, delay req 60.00
2014-07-28 13:06:18.1481 [PID=2086 ] Scheduler ran 6.311 seconds

Basically you're run out of allocated disk space for Boinc, of the 100 Gigabytes you're got allocated you're only got ~86 Megabytes left,
You could be using almost all of those 100 Gigs, or you could be falling foul of the Disk: use at most 50% of total preference, you could try upping the 100 Gigs to 110 Gigs, or you could try upping the % total up to 60%

Unlike most other projects where a Wu gets a single data file, Einstein uses a set of data files for multiple Wu's (at least on some of their searches), that means your host will accumulate data files, and you'll get sent tasks for the data files that you already have,
it does mean you may have lots of disk space tried up with datafiles awaiting for the server to tell Boinc to delete them, you could reset the project, but then you'd need to get fresh datasets, and that'll likely increase your downloads by a large amount,
best to increase the space allocated for now.

Claggy
4) Message boards : Cruncher's Corner : Only 2 threads running instead of 8 (Message 132533)
Posted 26 days ago by Claggy
NB.I am using Windows 7. I do not see any event log similar to your dump. I see many entries in the control panel, but not that particular one.

That is the server log for your last contact, the link's on the righthand side:

Computers belonging to Thierry Lombry

Claggy
5) Message boards : Problems and Bug Reports : Download stalls at same percentage (Message 132528)
Posted 26 days ago by Claggy
ARM is another computer I use without any issues....

this post is regarding non-ARM system. One of the other 5 claggy.

I know, that post wasn't replying to you, it was replying to Logforme who was making faulty assumptions.

Claggy
6) Message boards : Problems and Bug Reports : Download stalls at same percentage (Message 132519)
Posted 26 days ago by Claggy
Could it be that "plex" is the Arm processor one? I know that S6CasA tasks are very large on disk (it is using 20GB on my Win7 machine) and the Arm machine reports "only" 3.5GB free disk.

? The only applications the project has for Arm processors is for BRP, there is no CASA application for Arm at present.

Einsteinathome applications

Claggy
7) Message boards : Cruncher's Corner : Only 2 threads running instead of 8 (Message 132518)
Posted 26 days ago by Claggy
Your last contact:

http://einstein.phys.uwm.edu/host_sched_logs/10689/10689021

2014-07-27 21:55:03.7922 [PID=7678] Request: [USER#xxxxx] [HOST#10689021] [IP xxx.xxx.xxx.26] client 7.2.42
2014-07-27 21:55:03.8076 [PID=7678 ] [send] effective_ncpus 7 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2014-07-27 21:55:03.8076 [PID=7678 ] [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2014-07-27 21:55:03.8076 [PID=7678 ] [send] Not using matchmaker scheduling; Not using EDF sim
2014-07-27 21:55:03.8076 [PID=7678 ] [send] CPU: req 303660.00 sec, 7.00 instances; est delay 0.00
2014-07-27 21:55:03.8076 [PID=7678 ] [send] CUDA: req 0.00 sec, 0.00 instances; est delay 0.00
2014-07-27 21:55:03.8077 [PID=7678 ] [send] work_req_seconds: 303660.00 secs
2014-07-27 21:55:03.8077 [PID=7678 ] [send] available disk 0.09 GB, work_buf_min 0
2014-07-27 21:55:03.8077 [PID=7678 ] [send] active_frac 0.999945 on_frac 0.998163 DCF 0.898377
2014-07-27 21:55:03.8087 [PID=7678 ] [send] [HOST#10689021] is reliable
2014-07-27 21:55:03.8088 [PID=7678 ] [send] set_trust: random choice for error rate 0.000010: yes
2014-07-27 21:55:03.8090 [PID=7678 ] [mixed] sending non-locality work first (0.5126)
2014-07-27 21:55:03.8241 [PID=7678 ] [version] Checking plan class 'opencl-intel_gpu'
2014-07-27 21:55:03.8253 [PID=7678 ] [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2014-07-27 21:55:03.8254 [PID=7678 ] [version] numerical Windows version: 601760100 (Microsoft Windows 7 Home Premium x64 Edition, Service Pack 1, (06.01.7601.00))
2014-07-27 21:55:03.8254 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8254 [PID=7678 ] [version] No Intel GPU devices found
2014-07-27 21:55:03.8254 [PID=7678 ] [version] Checking plan class 'opencl-intel_gpu'
2014-07-27 21:55:03.8255 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8255 [PID=7678 ] [version] No Intel GPU devices found
2014-07-27 21:55:03.8255 [PID=7678 ] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2014-07-27 21:55:03.8255 [PID=7678 ] [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2014-07-27 21:55:03.8255 [PID=7678 ] [version] Checking plan class 'BRP5-opencl-ati'
2014-07-27 21:55:03.8255 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8255 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8255 [PID=7678 ] [version] Checking plan class 'BRP5-cuda32'
2014-07-27 21:55:03.8255 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8255 [PID=7678 ] [version] driver version required max: -29053, supplied: 33788
2014-07-27 21:55:03.8255 [PID=7678 ] [version] Checking plan class 'BRP5-cuda32-nv301'
2014-07-27 21:55:03.8256 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8256 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11
2014-07-27 21:55:03.8256 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:03.8256 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 139 for einsteinbinary_BRP5 (BRP5-cuda32-nv301)
2014-07-27 21:55:03.8256 [PID=7678 ] [version] Checking plan class 'BRP5-opencl-ati'
2014-07-27 21:55:03.8256 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8256 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8256 [PID=7678 ] [version] no app version available: APP#23 (einsteinbinary_BRP5) PLATFORM#9 2014-07-27 21:55:03.0995 [PID=7717 ] SCHEDULER_REQUEST::parse(): unrecognized: <allow_multiple_clients>0</allow_multiple_clients>


2014-07-27 21:55:03.8256 [PID=7678 ] [version] no app version available: APP#23 (einsteinbinary_BRP5) PLATFORM#2 (windows_intelx86) min_version 0
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Checking plan class 'BRP4G-opencl-ati'
2014-07-27 21:55:03.8257 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8257 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Checking plan class 'BRP4G-cuda32'
2014-07-27 21:55:03.8257 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8257 [PID=7678 ] [version] driver version required max: -29053, supplied: 33788
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Checking plan class 'BRP4G-cuda32-nv301'
2014-07-27 21:55:03.8257 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11
2014-07-27 21:55:03.8257 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 139 for einsteinbinary_BRP4G (BRP4G-cuda32-nv301)
2014-07-27 21:55:03.8257 [PID=7678 ] [version] Checking plan class 'BRP4G-opencl-ati'
2014-07-27 21:55:03.8257 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2014-07-27 21:55:03.8258 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8258 [PID=7678 ] [version] no app version available: APP#25 (einsteinbinary_BRP4G) PLATFORM#9 (windows_x86_64) min_version 0
2014-07-27 21:55:03.8258 [PID=7678 ] [version] no app version available: APP#25 (einsteinbinary_BRP4G) PLATFORM#2 (windows_intelx86) min_version 0
2014-07-27 21:55:03.8258 [PID=7678 ] [version] Checking plan class 'FGRPopencl-ati'
2014-07-27 21:55:03.8258 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2014-07-27 21:55:03.8258 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8258 [PID=7678 ] [version] Checking plan class 'FGRPopencl-nvidia'
2014-07-27 21:55:03.8258 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2014-07-27 21:55:03.8258 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11
2014-07-27 21:55:03.8258 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 111 for hsgamma_FGRP3 (FGRPopencl-nvidia)
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Checking plan class 'FGRPopencl-ati'
2014-07-27 21:55:03.8259 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2014-07-27 21:55:03.8259 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Checking plan class 'FGRPopencl-nvidia'
2014-07-27 21:55:03.8259 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11
2014-07-27 21:55:03.8259 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 111 for hsgamma_FGRP3 (FGRPopencl-nvidia)
2014-07-27 21:55:03.8259 [PID=7678 ] [version] Checking plan class 'FGRPSSE'
2014-07-27 21:55:03.8259 [PID=7678 ] [version] project prefs setting 'also_run_cpu' (1.000000) prevents using plan class.
2014-07-27 21:55:03.8259 [PID=7678 ] [version] no app version available: APP#26 (hsgamma_FGRP3) PLATFORM#9 (windows_x86_64) min_version 0
2014-07-27 21:55:03.8259 [PID=7678 ] [version] no app version available: APP#26 (hsgamma_FGRP3) PLATFORM#2 (windows_intelx86) min_version 0
2014-07-27 21:55:03.8298 [PID=7678 ] [mixed] sending locality work second
2014-07-27 21:55:09.6514 [PID=7678 ] [version] Checking plan class 'GWopencl-nvidia-Beta'
2014-07-27 21:55:09.6514 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_gw': 1.000000
2014-07-27 21:55:09.6514 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11


2014-07-27 21:55:09.6514 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:09.6515 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 108 for einstein_S6CasA (GWopencl-nvidia-Beta)
2014-07-27 21:55:09.6515 [PID=7678 ] [version] Checking plan class 'SSE2'
2014-07-27 21:55:09.6515 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:09.6515 [PID=7678 ] [version] Checking plan class 'SSE2-Beta'
2014-07-27 21:55:09.6515 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:09.6515 [PID=7678 ] [version] Checking plan class 'GWopencl-ati-Beta'
2014-07-27 21:55:09.6515 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_gw': 1.000000
2014-07-27 21:55:09.6515 [PID=7678 ] [version] No ATI devices found
2014-07-27 21:55:09.6515 [PID=7678 ] [version] Checking plan class 'GWopencl-nvidia-Beta'
2014-07-27 21:55:09.6516 [PID=7678 ] [version] parsed project prefs setting 'gpu_util_gw': 1.000000
2014-07-27 21:55:09.6516 [PID=7678 ] [version] Peak flops supplied: 1.38816e+11
2014-07-27 21:55:09.6516 [PID=7678 ] [version] plan class ok
2014-07-27 21:55:09.6516 [PID=7678 ] [version] Don't need CUDA jobs, skipping version 108 for einstein_S6CasA (GWopencl-nvidia-Beta)
2014-07-27 21:55:09.6516 [PID=7678 ] [version] Best version of app einstein_S6CasA is ID 567 (6.23 GFLOPS)
2014-07-27 21:55:09.6525 [PID=7678 ] [send] stopping work search - insufficient disk space
2014-07-27 21:55:09.6525 [PID=7678 ] [send] stopping work search - insufficient disk space
2014-07-27 21:55:09.6525 [PID=7678 ] [send] stopping work search - insufficient disk space
2014-07-27 21:55:09.6525 [PID=7678 ] [send] stopping work search - insufficient disk space

2014-07-27 21:55:09.6569 [PID=7678 ] [debug] [HOST#10689021] MSG(high) No work sent
2014-07-27 21:55:09.6569 [PID=7678 ] [debug] [HOST#10689021] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/10689/10689021
2014-07-27 21:55:09.6569 [PID=7678 ] [debug] [HOST#10689021] MSG(high) Gravitational Wave S6 Directed Search (CasA) needs 6.64MB more disk space. You currently have 88.72 MB available and it needs 95.37 MB.
2014-07-27 21:55:09.6569 [PID=7678 ] Sending reply to [HOST#10689021]: 0 results, delay req 60.00
2014-07-27 21:55:09.6572 [PID=7678 ] Scheduler ran 5.872 seconds


You can't get FGRP CPU work because you don't have 'Run CPU versions of applications for which GPU versions are available' set to Yes in your project preferences,

And you can't get S6CasA work because your've run out of disk space.

Claggy
8) Message boards : Problems and Bug Reports : Nvidia cuda driver (Message 132491)
Posted 28 days ago by Claggy
Thank you. I just turned cuda on and am waiting the get work.

Make sure you have 'Use NVIDIA GPU' set to Yes in your Einstein project preferences, at present neither of your hosts are asking for Nvidia GPU work,
You'll might also want to set 'Binary Radio Pulsar Search (Arecibo, GPU)' and 'Binary Radio Pulsar Search (Perseus Arm Survey)' to Yes too.

Claggy
9) Message boards : Problems and Bug Reports : Nvidia cuda driver (Message 132488)
Posted 28 days ago by Claggy
There isn't a separate cuda driver, it is all packaged together:

http://www.nvidia.com/Download/index.aspx?lang=en-us

Claggy
10) Message boards : Problems and Bug Reports : Validate error - What this really means! (Message 132483)
Posted 28 days ago by Claggy
------> Used in total: 299 MB (1750 MB free / 2049 MB total) -> Used by this application (assuming a single GPU task): 0 MB
[15:01:26][3632][INFO ] Using CUDA device #0 "GeForce GTX 770" (0 CUDA cores / 0.00 GFLOPS)
[15:01:26][3632][INFO ] Version of installed CUDA driver: 6000
[15:01:26][3632][INFO ] Version of CUDA driver API used: 3020


I wouldn't worry about that, the Cuda 3.2 api doesn't know about how GTX 770's are made up,
Arvid Almstrom's GTX780's top Nvidia host also doesn't report the number of Cuda cores or GFLOPS :

http://einstein.phys.uwm.edu/show_host_detail.php?hostid=6216490

[18:11:04][11780][INFO ] CUDA global memory status (initial GPU state, including context):
------> Used in total: 675 MB (2399 MB free / 3074 MB total) -> Used by this application (assuming a single GPU task): 0 MB
[18:11:04][11780][INFO ] Using CUDA device #1 "GeForce GTX 780" (0 CUDA cores / 0.00 GFLOPS)
[18:11:04][11780][INFO ] Version of installed CUDA driver: 6000
[18:11:04][11780][INFO ] Version of CUDA driver API used: 3020

and on my GT650M:

[18:30:27][5600][INFO ] CUDA global memory status (initial GPU state, including context):
------> Used in total: 71 MB (1978 MB free / 2049 MB total) -> Used by this application (assuming a single GPU task): 0 MB
[18:30:27][5600][INFO ] Using CUDA device #0 "GeForce GT 650M" (0 CUDA cores / 0.00 GFLOPS)
[18:30:27][5600][INFO ] Version of installed CUDA driver: 6050
[18:30:27][5600][INFO ] Version of CUDA driver API used: 3020

Claggy


Next 10

Home · Your account · Message boards

This material is based upon work supported by the National Science Foundation (NSF) under Grants PHY-1104902, PHY-1104617 and PHY-1105572 and by the Max Planck Gesellschaft (MPG). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the NSF or the MPG.

Copyright © 2014 Bruce Allen