Susanne:
I'd like to respond to your comments.
I believe that the reference to "network partitions" refers to
the fact that in a distributed environment, unlike a single machine
environment, we cannot be sure that messages will be delivered: a network
failure can result in any message being lost. Thus, a job submission may
not receive a response, and in that case, we cannot know whether the job
was submitted (i.e., the request got through, but the response was lost)
or not (i.e., the request was lost).
One convenient way of dealing with this problem is to allow users to
associate a "unique job id" with a job request. A scheduler
that receives a second or subsequent submission with the same jobid
should simply return the response it provided to the first request
received.
It's true that a user can achieve a similar effect by searching for the
submitted job in the scheduler queue. However, this approach is more
complex to do, and also suffers from the problem that the job might have
already completed and thus can't be found that way.
In our recently circulated ESI specification, we proposed an optional
"unique job id" field in JSDL as a way of addressing this
requirement. This notion was discussed on a BES call, and people seemed
sympathetic to the idea.
Ian.
At 12:20 PM 4/28/2006 -0400, Balle, Susanne wrote:
Marvin,
Enclosed find the remaining of my comments:
Page 5. (top paragraph) I think I know what you mean by "with
the
ambiguity of distinguishing between scheduler crashes and network
partitions ". "scheduler crashes" is obvious. I am
assuming that by
"network partitions" you are inferring that various
sub-networks are
going to have different response time which will have an effect on
the
time it takes to deliver a call-back message.
Reading further along in the same paragraph I am now not sure I
know
what you mean by "network partitions".
Page 5. Section 3.3
The topic of this section is clear (described in the first line of
the
paragraph) but of the section is a little confusing.
"possibility that a client cannot directly tell whether its
job
submission request has been successful ..." --> Do we expect the
client
to re-submit the job if the submission failed or do we expect users
to
inspect that their job has in fact been submitted and resubmit if
needed? I am wondering if we assume the later if that wouldn't result
in
users re-launching their jobs several time if they do not see their
job
listed in some state when pulling the job scheduler for the state
of
their job?
I guess I do not understand why so much emphasis is put on the
"At-Most-Once" or "Exactly-Once".
Can't the client poll the Job scheduler and ask the JS for a list
of
jobs queued, running, terminated, failed, etc.? It might be useful
for
the client to be able to submit jobs with a special keyword like
JOB_SUBMITTED_BY since that would reduce the list it gets back. It
would
be nice if the value for the keyword was a unique identifier but
doesn't
have to be. Most schedulers allows you to name or associate a group
to
programs so that feature could be used as special keyword.
Page 6. Section 3.4
General question: Are you taking into account that user
applications
will require different software?
1. For example if my executable is compiled for Linux, Intel
platform
then I would like to run it on a Linux,Intel system and not a
Linux,AMD
system.
2. Are you assuming that the program will be compiled on the fly on
the
allocated system? or pre-compiled and then staged?
I agree that staging the data is going to be an interesting
topic.
All this is probably out-of-band for the HPC JS Profile but should
be
considered somewhere. I am sure it is I just don't know where.
I like the section on virtual machines and think that they will be
used
more and more in the future.
Page 7. Extended Resource Features
The second approach (arbitrary resource types ...) is the only one
that
make sense to me since that approach is extensible. I believe that
Moab
is implementing this approach as well.
Page 8. Extended Client/System Administrator Operations
Are you assuming that System Administrators will be able to perform
sys
admin operations on somebody else's system? I don't think that is
right.
You mention suspend-resume. Are you thinking of suspending a job
running
across several clusters that are in different organizations? Or
just
suspending a job on a single cluster/server?
Again I am trying to figure out how this fit in with "One
important
aspect, is that the individual clusters should remain under the
control
of their local system administrator and/or of their local policies".
I believe that suspend-resume is a JS operation or an operation to
be
performed by the local sys admin, NOT by remote sys admins.
If we are now talking about a meta-scheduler then yes it makes sense.
In
the case of a meta-scheduler it might take over the individual JS
and
schedule jobs base on its own policies, on its job reservation
system,
etc. In this case I look at it as we have one deciding entity (the
meta-scheduler) and several "slaves". Moab and Maui are the
only
meta-scheduler I an familiar with and they do take over the
scheduling
decisions/node allocations/etc and just submit jobs to the local
job
schedulers.
This does of course assume that the local system administrators
have
agreed on a schedule when their cluster is shared within this
greater
infrastructure. This is a different approach than having jobs
passed
onto their local scheduler and run on their systems.
This just seems to be a different approach from the one that is taken
in
this paper.
I might be wrong. If I am please educate me.
Page 9. Section 3.10
Don't forget UPC (Unified Parallel C: http://upc.nersc.gov/). This
parallel programming paradigm is getting more and more interest
from
several communities.
We'll need to provide support for UPC as well.
Page 10. Section 3.13
A meta-scheduler approach that make sense to me is to allow
developers
to submit their job to their local cluster using their
"favorite"
scheduler commands and then have the meta-scheduler load-balance
the
work and forward the job to another system/cluster if needed. Moab
from
cluster resources support this approach even if the clusters have
different JSs. They have a list of supported JS such as LSF,
PBSpro,
SLURM, etc. and they can "translate" one JS's commands into
another
within that supported set.
Page 11. SLURM is missing.
Let me know what you think,
Regards
Susanne
---------------------------------------------------------------
Susanne M. Balle,
Hewlett-Packard
High Performance Computing R&D Organization
110 Spit Brook Road
Nashua, NH 03062
Phone: 603-884-7732
Fax: 603-884-0630
Susanne.Balle@hp.com
_______________________________________________________________
Ian Foster, Director, Computation Institute
Argonne National Laboratory & University of Chicago
Argonne: MCS/221, 9700 S. Cass Ave, Argonne, IL 60439
Chicago: Rm 405, 5640 S. Ellis Ave, Chicago, IL 60637
Tel: +1 630 252 4619. Web:
www.ci.uchicago.edu.
Globus Alliance:
www.globus.org.