On 06/12/2016 04:01 PM, Andrew Savchenko wrote:
On Thu, 2 Jun 2016 10:42:03 -0500 james wrote:
On 06/01/2016 06:20 PM, Justin Bronder wrote:
> Due to a lack of time and the fact I don't use any of these packages
> anymore, they are all up for grabs.
>
> - media-gfx/openmesh [no project]
> - sys-cluster/ganglia [cluster]
> - sys-cluster/ganglia-web [cluster]
> - sys-cluster/torque [cluster]
> - sys-cluster/munge [cluster] dependency of sys-cluster/torque
> - sys-cluster/mpe2 [cluster]
>
> Also, if there's anyone out there using the science overlay and empi
> who's feeling motivated, that work still needs a champion to get it
> into the main tree. If not, I'll probably drop it in a few months
> and open openmpi and mpich2 to project maintenance as well. I
> haven't been involved in HPC for over a decade now, it's time to
> pass the torch.
Hello Justin,
I've been working on cluster ebuilds for a while (Apache Mesos, spark,
etc). I'm willing to proxy maintain these except torque. Assuming there
are no users of torque on gentoo (bgo seems inactive...it's dead; how
would I know?).
Hey, don't be too quick on your judgement :) We are using torque
(and maui) on three HPC setups, though we had to patch torque for
various features.
Ian picked up torque::
"I can step up and maintain or co-maintain sys-cluster/torque , I use
it at work and have contributed to it in the past."
Which was written on 6/2/16 by a...@gentoo.org.
By the way, what scheduler/resource manager are you using for HPC?
I'm more focused on small clusters and the ability to 'reboot' identical
hardware into a variety of HPC or container cluster configurations. My
immediate goal is fast and small clusters to run 'cluster-benchmarks'
and compare different cluster offerings, for a particular problem. The
emphasis is the hardware which is identical, so the cluster codes are
the only difference. NO elaborate Schedulers or Framework schedulers,
just testing everything, atm.
That said, you may want to look at Mesos, as that cluster OS, it
supports a myriad of Schedulers and Framework schedulers. It fact it
encourages custom scheduler development.
http://mesos.apache.org/
http://mesos.apache.org/documentation/latest/app-framework-development-guide/
slurm ebuild seems to be even in worse state than torque one (2.6.3
is present, 16.05.0 is the latest upstream).
I do not know why more gentoo devs have not been bitten by the
sys-cluster bug. Slurm is still widely used. SchedMD is hiring.
> Best regards,
> Andrew Savchenko
> Best regards,
> Andrew Savchenko
James
[1]
http://mesos.apache.org/documentation/latest/app-framework-development-guide/
