Posts by Henk Haneveld

1) Message boards : Number crunching : Disk full (Message 640)
Posted 1 May 2021 by Henk Haneveld
Post:
01/05/2021 13:44:55 | BOINC@TACC | [error] Error reported by file upload server: Server is out of disk space
2) Message boards : News : BOINC@TACC back in production (Message 602)
Posted 11 Dec 2020 by Henk Haneveld
Post:
You claim that a data set of several 100 GB is a problem. Have a look at the LHC project, they have that kind of work on a daily basis without problem.

You want to process that kind of work, then put it in the work queue and it will be done.
3) Message boards : News : Supporting SARS-COV2 related research (Message 373)
Posted 17 Apr 2020 by Henk Haneveld
Post:
I run version 5.2.38
4) Message boards : News : Supporting SARS-COV2 related research (Message 307)
Posted 9 Apr 2020 by Henk Haneveld
Post:
Hello,

Thank you to you all for bringing it to our attention. We have reduced the number of jobs run in a client as 5 per CPU core. i.e. : a processor with 4 cores can get maximum 20 jobs at the same time. There will be validation run for the same number of jobs. Hence the host would be running 40 jobs in this example.

Sincerely,
the BOINC@TACC development team

The change to 5 per core looks to work nicely. What I don't understand is how you get the number of 40 jobs

What I do notice is that sometimes I get both the _0 en _1 version of the same job.
If you want solid validation they should be send to separate hosts and you should check your settings to fix that..
If validation is not that important then there is no point for sending out 2 versions of the same job
5) Message boards : News : Supporting SARS-COV2 related research (Message 294)
Posted 8 Apr 2020 by Henk Haneveld
Post:
Admin, when you do that you may want to reconsider the settings to use.
In an other post you gave a fixed number of max 256 results per host.
It would be beter to use a fixed number per core.
Example: 2 per core on a 128 core will also give 256 results but for a host with only 4 cores this wil be a max of just 8 results.
This will give a more even distribution based on the processing capacity of the host.

Henk

Greetings - thanks for letting us know. We will revisit the settings again today.

Best Regards,
The BOINC@TACC Team

Hello,

As there will be a constant flow of work for the next few weeks, if a host is done with processing a set of jobs, the host can always fetch additional work-units. There will be several thousand jobs submitted daily for the next few days.
We will watch the job turn around time with the current setting, and we can always adjust the settings for limiting the number of work-units on a computer at a given point in time. As of now, each host will be able to have 256 active tasks from the queue at any instance of time. If they finish processing for example128 tasks, they can always get additional 128 tasks from the queue. Any volunteer can continue requesting jobs from the queue once they finish processing jobs and stay below the limit.

Sincerely,
the BOINC@TACC development team


Just wondering if that limitation is working as expected. My oldish laptop downloaded about 700 tasks today and over 600 are still waiting to start. Thats enough work for the next days while other computers may be idle.
6) Message boards : News : Supporting SARS-COV2 related research (Message 271)
Posted 7 Apr 2020 by Henk Haneveld
Post:
Admin, I noticed that the project executes a project server call every minute. Why?

I can not find a usefull reason for it. You create a high load on your server without any gain.
7) Message boards : News : Supporting SARS-COV2 related research (Message 251)
Posted 5 Apr 2020 by Henk Haneveld
Post:
Hello Admin,

Altough you have send out several thousands of results I have got only a few.
After some checking I found that some users have got a large numbers of results in their queue.

May I suggest that you implement the max work in progress parameter.
It will give all users a fair change on work and it will lead to a faster turnaround time of work you send out.
8) Message boards : Number crunching : Validation is wrong (Message 95)
Posted 3 May 2019 by Henk Haneveld
Post:
I got a batch of work but when I checked if they got validated I found that both results for the same workunit had been send to my host and validated against each other.

If validation is important then results should be send to seperate hosts, if it is not there is no need to send out 2 results.
9) Message boards : Number crunching : Data file cleanup (Message 69)
Posted 16 Mar 2019 by Henk Haneveld
Post:
I noticed that after the results are done and returned I still have a lot of data files in the project directory all ending with ....tar.manual.gz

Are they needed for future work or can they be deleted.
And if they can be deleted, the project should do this automaticly.