Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 2
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1473 times and has 1 reply Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Job sizing to suite target platform.

There was discussion, even berkeley server development work to size jobs according the performance of the various platforms even down to cpu power dynamic sizing. For a while been seeing 140 jobs being packed into android fahv tasks, some taking 18+ hours. The longer they run, the greater the statistical chance they're not going to make it to the end with all that great stability the android app packs. At wuprop we see some logging 61 hours runtime. Not seen any fahv on regular nodes for a while, the question being how many jobs are in those? If same, we've got the answer, no progress in this department, should we say 'either'?

Help me out here, why is there this feeling the glass just does not seem to get even half full?
[Aug 4, 2014 5:27:03 PM]   Link   Report threatening or abusive post: please login first  Go to top 
seippel
Former World Community Grid Tech
Joined: Apr 16, 2009
Post Count: 392
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Job sizing to suite target platform.

The fahv tasks are currently the same between Android and non-Android. There is a current development task (that is actively being worked) to add the abiility to create variable sized work units which would allow smaller work units for Android and larger ones for non-Android.

Seippel
[Aug 7, 2014 3:46:49 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread