How do you configure so that boinc processes 1 task (1 by 1), finishes this, and requests another task?

TIMID
TIMID
Joined: 17 Jan 22
Posts: 3
Credit: 22,869
RAC: 50
Topic 226856

Hi, actually I setup on website as images (link). But when update or project get task, sent me 6+ task to process on parallel, I need run 2 or 1 on parallel.

How do it?.

https://res.cloudinary.com/www-solucionespetabyte-com/image/upload/v1643214950/Untitled-2_wrje7k.jpg

https://res.cloudinary.com/www-solucionespetabyte-com/image/upload/v1643214960/Untitled-1_mykidc.jpg

https://res.cloudinary.com/www-solucionespetabyte-com/image/upload/v1643214894/Untitled_g26uis.png

 

Specs

1/26/2022 9:22:49 AM |  | cc_config.xml not found - using defaults
1/26/2022 9:22:50 AM |  | Starting BOINC client version 7.16.20 for windows_x86_64
1/26/2022 9:22:50 AM |  | log flags: file_xfer, sched_ops, task
1/26/2022 9:22:50 AM |  | Libraries: libcurl/7.47.1 OpenSSL/1.0.2s zlib/1.2.8
1/26/2022 9:22:50 AM |  | Data directory: D:\ProgramData\BOINC
1/26/2022 9:22:50 AM |  | Running under account TIMID
1/26/2022 9:22:50 AM |  | CUDA: NVIDIA GPU 0: NVIDIA GeForce GTX 1650 SUPER (driver version 497.29, CUDA version 11.5, compute capability 7.5, 4096MB, 3310MB available, 4416 GFLOPS peak)
1/26/2022 9:22:50 AM |  | OpenCL: NVIDIA GPU 0: NVIDIA GeForce GTX 1650 SUPER (driver version 497.29, device version OpenCL 3.0 CUDA, 4096MB, 3310MB available, 4416 GFLOPS peak)
1/26/2022 9:22:55 AM |  | Windows processor group 0: 12 processors
1/26/2022 9:22:55 AM |  | Host name: TIMID
1/26/2022 9:22:55 AM |  | Processor: 12 GenuineIntel Intel(R) Core(TM) i5-10400 CPU @ 2.90GHz [Family 6 Model 165 Stepping 3]
1/26/2022 9:22:55 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 tm2 pbe fsgsbase bmi1 smep bmi2
1/26/2022 9:22:55 AM |  | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.19044.00)
1/26/2022 9:22:55 AM |  | Memory: 7.86 GB physical, 12.98 GB virtual
1/26/2022 9:22:55 AM |  | Disk: 1.82 TB total, 436.66 GB free
1/26/2022 9:22:55 AM |  | Local time is UTC -5 hours
1/26/2022 9:22:55 AM |  | WSL detected:
1/26/2022 9:22:55 AM |  | [kali-linux] (default): Linux Kali (Kali GNU/Linux Rolling [5.10.16.3-microsoft-standard-WSL2])
1/26/2022 9:22:55 AM |  | VirtualBox version: 6.1.30
 

TIMID

GWGeorge007
GWGeorge007
Joined: 8 Jan 18
Posts: 730
Credit: 1,254,708,779
RAC: 4,261,961

Hi TIMID!  Welcome to

Hi TIMID!  Welcome to Einstein@Home!

I'll try to help you given that you are relatively new to Einstein.

Your event log file shows:


1/26/2022 9:22:49 AM |  | cc_config.xml not found - using defaults


You will need to make a <cc_config.xml> file first, then place in a hidden directory which may not be shown.  If so, you may need to make Windows show hidden files before you place it in there.  Save it in your BOINC Data directory or project directory.

Study this so you have an understanding of what to do:


https://boinc.berkeley.edu/wiki/Client_configuration


In your <cc_config.xml> file, I think if you put this line item in place it should give you a single file every time you finish with one.


<ncpus>1</ncpus>


If you are running Einstein tasks for the CPU and want only one at a time, this should work.  If you want 2 or more, then just change the number to the number that you want.  If you are using the GPU for Einstein, just substitute "g" for the "c", and do the same thing.

I hope this helps.  Let us know if it works for you.

.....[EDIT].....

I forgot to add... After doing the above, you'll need to restart BOINC.

George

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5,542
Credit: 76,581,465,829
RAC: 64,970,172

TIMID wrote:... 6+ task to

TIMID wrote:

... 6+ task to process on parallel, I need run 2 or 1 on parallel.

How do it?.

First of all, welcome to Einstein@Home.  We trust you will enjoy your time here.

You do NOT need to create either cc_config.xml OR app_config.xml to limit the number of "parallel" tasks.  The default settings are fine.  Those files are not necessary unless you need something much more advanced than what you have requested.

If you want this to apply to all machines under your account, you can make the change on the website.  If you want it to apply to a single machine (and override website settings) you can make the change locally in BOINC Manager (Advanced view).

For a global change, go to your computing preferences on the website.  You will see "Preference set:" which should show as "generic" if you haven't changed this to a specific "location" (eg. home/work/school) that is different.  The default preference set uses the "generic" location if not previously changed.

Below that, there is a dropdown called "ADVANCED SETTINGS".  If you open that you will see a whole host of individual settings you can adjust.

Scroll down in the "Processor usage" section and the second last item says "Use at most ... 100% of the processors".  Change the value to what you need.  If you normally had 6 running tasks with the default 100%, then setting that to 16% would give you just 1 and 33% would give you 2.  Then go to the very bottom of the page and click the "Save Changes" link.

After making that change, you can select the Einstein project in BOINC Manager locally (on the Projects tab) and click "Update" so that your BOINC client can contact the project and become aware of the change immediately, rather than having to wait for a normal contact that might not happen for a while.

Rather than making a global change, if you wanted this to apply to just the one machine, then open BOINC Manager on that machine in the 'Advanced view' mode and select "Options -> Computing preferences" where the very first item under "Usage limits" allows you to make this change locally.  Just remember that you will now have local settings that will override Website settings for that machine.  There is a button called "Use web prefs" that will allow you to go back to website settings if you ever need to.

Cheers,
Gary.

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 2,133
Credit: 5,332,571,173
RAC: 19,222,822

There's a couple of different

There's a couple of different ways to achieve what you want.

I like using project_max_concurrent in an app_config.xml file.

The link for the docs for client configuration examples was given.

 

TIMID
TIMID
Joined: 17 Jan 22
Posts: 3
Credit: 22,869
RAC: 50

Thanks let me try it. 

Thanks let me try it. 

TIMID

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 2,133
Credit: 5,332,571,173
RAC: 19,222,822

I also think you should

I also think you should reduce your cache requests.  You still haven't returned the 1st tasks you received from several days ago.  Your cache is currently set for 1 day which is too much based on the little progress you have achieved. This is what you are requesting now.

work_buf_min 86400

I would reduce that to 0.1 or maybe 0.25 days worth of work.

 

TIMID
TIMID
Joined: 17 Jan 22
Posts: 3
Credit: 22,869
RAC: 50

thanks, It s working. I will

thanks, It s working. I will test this week. :)

TIMID

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.