Welcome to MilkyWay@home

Posts by nedmanjo

1) Message boards : Number crunching : warnings & errors (Message 68731)
Posted 14 May 2019 by Profilenedmanjo
Post:
Thanks so much for explaining. I appreciate the time you took to respond in detail and the knowledge you passed on.
2) Message boards : Number crunching : warnings & errors (Message 68724)
Posted 12 May 2019 by Profilenedmanjo
Post:
Toggled this line, <project_max_concurrent>x</project_max_concurrent>, about and changes in value are reflected in total # WU's running. This seems most useful as it limits the total number of WU's running.

<avg_ncpus>x</avg_ncpus> - the number of CPU instances used by the app version.

I'm thinking the CPU instances the same as cores. I have 2 12-core CPU's
If I'm running 6 WU's and set x to 6, does that mean, on average, 1 CPU (core) would be available per WU?
And if I set it to 12, on average, 2 CPU (core) would be available per WU?

<ngpus>x</ngpus> - the number of GPU instances used by the app

If nbody is a CPU task then what is the purpose to adding this line in the app_config file?

<cmdline>--nthreads x</cmdline>

What I could find on the BOINC message board.

avg_ncpus tells BOINC how many threads to reserve for the job scheduling
cmdline tells BOINC how many threads the application may use at maximum
Milkway Nbody tasks can run on any number of threads that you care to set for them, but do keep the numbers you use for avg_ncpus and nthreads equal.
3) Message boards : Number crunching : warnings & errors (Message 68721)
Posted 12 May 2019 by Profilenedmanjo
Post:
Also, looking at my client I have 37 WU's of nbody simulations running along with 2 GPU separation units, each at 0.5 CPU.

I have 24 physical cores and 24 virtual cores and am allowing 80% of the cores to be used. Would appear that each WU is running on a core?

I thought the app_config would assign multiple cores to a single task? Perhaps speeding up the computations?
4) Message boards : Number crunching : warnings & errors (Message 68720)
Posted 12 May 2019 by Profilenedmanjo
Post:
Thank you for responding.

Would prefer to correct rather than ignore but if the errors are harmless, and they do seem to be, I suppose just ignore it.

Is this syntax giving you 4 threads?
Always seen this as integer input:
<cmdline>--nthreads 4</cmdline>

Made the change but don't know how to confirm the thread count. Where would I look?
5) Message boards : Number crunching : warnings & errors (Message 68718)
Posted 12 May 2019 by Profilenedmanjo
Post:
Looking at the task output I'm seeing some errors & warnings and don't know how to address them.

<search_application> milkyway_separation 1.46 Windows x86 double OpenCL </search_application>
Reading preferences ended prematurely
BOINC GPU type suggests using OpenCL vendor 'Advanced Micro Devices, Inc.'
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Switching to Parameter File 'astronomy_parameters.txt'
<number_WUs> 4 </number_WUs>
<number_params_per_WU> 26 </number_params_per_WU>

and

C:\Users\jnedd\AppData\Local\Temp\\OCL8644T1.cl:183:72: warning: unknown attribute 'max_constant_size' ignored
__constant real* _ap_consts __attribute__((max_constant_size(18 * sizeof(real)))),
^
C:\Users\jnedd\AppData\Local\Temp\\OCL8644T1.cl:185:62: warning: unknown attribute 'max_constant_size' ignored
__constant SC* sc __attribute__((max_constant_size(NSTREAM * sizeof(SC)))),
^
C:\Users\jnedd\AppData\Local\Temp\\OCL8644T1.cl:186:67: warning: unknown attribute 'max_constant_size' ignored
__constant real* sg_dx __attribute__((max_constant_size(256 * sizeof(real)))),
^
3 warnings generated.

WU's for both milkway & milkyway_nbody are completing without error.

Couple of notes:

BOINC is installed to my E: drive, not C:

Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz [Family 6 Model 62 Stepping 4]
(48 processors)
AMD Radeon Vega Frontier Edition (16368MB) OpenCL: 2.0

App_Config is as follows:

<app_config>
<app>
<name>milkyway</name>
<max_concurrent>4</max_concurrent>
<gpu_versions>
<gpu_usage>0.50</gpu_usage>
<cpu_usage>0.50</cpu_usage>
</gpu_versions>
</app>
<app_version>
<app_name>milkyway_nbody</app_name>
<plan_class>mt</plan_class>
<avg_ncpus>4.0</avg_ncpus>
<cmdline>--nthreads4.0</cmdline>
</app_version>
</app_config>

Also, looking in the project directory I no longer see the GPU configuration xml file for milkyway_1.46_windows_x86_64__opencl_ati_101. Is this not supported?
6) Message boards : Number crunching : Profiles? (Message 67317)
Posted 9 Apr 2018 by Profilenedmanjo
Post:
Having the same issue. reCaptcha does not display on IE, Firefox or Edge but the reCaptcha does display on all of my other projects. Seems the issue is unique to Milkyway.




©2019 Astroinformatics Group