The Mesos offer allocation cycle should happen once every second by default
(--allocation_interval), but if Myriad declines an offer (or doesn't use
all the resource in an offer), then those resources will be filtered from
being offered to that framework for 5sec (by default). We could reduce that
Santosh, I get a lot of 2-3 containers. But I can only get 9-12 (Topped of
the cpu resources at 12 cores) containers if a task runs for more than 30
secs (preferably 60 secs), that's generally not an issue but I thought it
was worth putting on the list for general knowledge. It's also less of a
> After the patches it seems stable, I'm able to run multiple terasort/pi
> jobs and a few scalding jobs without difficulty.
Great work, Darin. Glad to see FGS is now stable.
>Noticed with jobs with short map tasks (8-12 secs), I rarely got more
> than two containers per node, I'm curious if I'm
>>> Noticed with jobs with short map tasks (8-12 secs), I rarely got morethan
>>>two containers per node, I'm curious if I'm not consuming resources
fast enough
Yes, we have noticed that too. I think because heartbeat between Mesos and RM
is 5 sec (I think) we may not get enough resources with
I've been running a number of tests on the Fine Grained scaling aspect on
Myriad. Here's a few notes:
1. After the patches it seems stable, I'm able to run multiple terasort/pi
jobs and a few scalding jobs without difficulty.
2. Noticed with jobs with short map tasks (8-12 secs), I rarely got