giraph-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jaeho Shin (JIRA)" <>
Subject [jira] [Commented] (GIRAPH-274) Jobs still failing due to tasks timeout during INPUT_SUPERSTEP
Date Thu, 02 Aug 2012 05:18:02 GMT


Jaeho Shin commented on GIRAPH-274:

Maybe I wasn't clear enough.  My intention of creating a separate thread is to rather make
sure progress() gets called even in those periods the main thread is busy in some long running
external code or I/O calls than prettifying the code.  With the current approach of instrumenting
progress() calls along the path, even if we do call it next to every normal statement, it'll
still get killed whenever Giraph calls any other code that takes more than 600s to return.
 It seems as we push Giraph to the limit, we are seeing some unexpected parts become extremely
slow.  We're currently trying to understand its behavior by profiling it.  I'll post here
if we find something interesting.
> Jobs still failing due to tasks timeout during INPUT_SUPERSTEP
> --------------------------------------------------------------
>                 Key: GIRAPH-274
>                 URL:
>             Project: Giraph
>          Issue Type: Bug
>    Affects Versions: 0.2.0
>            Reporter: Jaeho Shin
>            Assignee: Jaeho Shin
>             Fix For: 0.2.0
>         Attachments: GIRAPH-274.patch
> Even after GIRAPH-267, jobs were failing during INPUT_SUPERSTEP when some workers don't
get to reserve an input split, while others were loading vertices for a long time.  (related
to GIRAPH-246 and GIRAPH-267)

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


View raw message