<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">-- This comment is coming too late for the discussion on the FT call, but I'll leave it here for further clarification of the issue with SHRINK. --<div><br></div><div>To be precise it depends on the communication mode, reset or continue. However, such concerns are technicalities. The real problem with SHRINK is that it cannot work properly if you do not break all existing communicators, as the local view of several software layers can be inconsistent.<div><br></div><div>In the context of FT-MPI, where all communicators were marked as invalid (on which any subsequent MPI call was bound to fail) on any [asynchronous] detection of lost processes, this approach make sense. In the context of the current proposal and discussion in the forum, where the same communicators can be fixed this approach is in my view very difficult to defend.</div><div><br></div><div> george.<br><div><br></div><div><br><div><div>On Jan 24, 2012, at 15:09 , Josh Hursey wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">It was also mentioned to me that the SHRINK mode also forces all messages to be dropped when the communicator is repaired. Shrinking will likely shift the ranks of processes in the communicator (to fill in the gaps) so matching will need to be reset, and messages would need to be dropped/drained. So it would no longer be possible for an application re-enable collectives and still have messages in-flight across the repair. At least not without further clarification on this point.<div>
<br></div><div>-- Josh<br><br><div class="gmail_quote">On Tue, Jan 24, 2012 at 2:41 PM, Graham, Richard L. <span dir="ltr"><<a href="mailto:rlgraham@ornl.gov">rlgraham@ornl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I will re-iterate what I said before. While this may one mode that apps may want to use, it is not the only mode. In particular, this forces all ranks in a communicator to know about the change, even if they have implemented a "local" algorithm that does not need to know about all failures.<br>
<br>
Rich<br>
<div class="im"><br>
On Jan 24, 2012, at 2:09 PM, Sur, Sayantan wrote:<br>
<br>
Hi Josh,<br>
<br>
Thanks for the crisp characterization of the proposal I was making. It is correct. I was naturally thinking of the SHRINK mode, since it involves least number of changes to MPI standard itself. Folks at the forum also had similar thoughts (e.g. why does MPI_Comm_size() still return count including failed procs).<br>
<br>
Cf. <a href="http://www.netlib.org/utk/people/JackDongarra/PAPERS/isc2004-FT-MPI.pdf" target="_blank">http://www.netlib.org/utk/people/JackDongarra/PAPERS/isc2004-FT-MPI.pdf</a><br>
<br>
“4.2 FTMPI_COMM_MODE_SHRINK<br>
<br>
In this communicator mode, the ranks of MPI processes before and after recovery might change, as well as the size of MPI COMM WORLD does change. The appealing part of this communicator mode however is, that all functions specified in MPI-1 and MPI-2 are still valid without any further modification, since groups and communicators do not have wholes (sic) and blank processes.”<br>
<br>
We can discuss further tomorrow as to whether we could go with SHRINK mode (simplifying the proposal). From what I read in the paper, they report being able to convert fault-tolerant master/worker to use both modes.<br>
<br>
Thanks!<br>
<br>
===<br>
Sayantan Sur, Ph.D.<br>
Intel Corp.<br>
<br>
</div>From: <a href="mailto:mpi3-ft-bounces@lists.mpi-forum.org">mpi3-ft-bounces@lists.mpi-forum.org</a><mailto:<a href="mailto:mpi3-ft-bounces@lists.mpi-forum.org">mpi3-ft-bounces@lists.mpi-forum.org</a>> [mailto:<a href="mailto:mpi3-ft-bounces@lists.mpi-forum.org">mpi3-ft-bounces@lists.mpi-forum.org</a>] On Behalf Of Josh Hursey<br>
<div class="im">Sent: Tuesday, January 24, 2012 10:39 AM<br>
To: MPI 3.0 Fault Tolerance and Dynamic Process Control working Group<br>
Subject: [Mpi3-ft] Choosing a BLANK or SHRINK model for the RTS proposal<br>
<br>
First let me say that I greatly appreciate the effort of Sayantan and others to push us towards considering alternative techniques, and stimulating discussion about design decisions. This is exactly the type of discussion that needs to occur, and the working group is the most appropriate place to have it.<br>
<br>
<br>
One of the core suggestions of Sayantan's proposal is the switch from (using FT-MPI's language) a model like BLANK to a model like SHRINK. I think many of the other semantics are derived from this core shift, so we should probably focus the discussion on this point earlier in our conversation.<br>
<br>
<br>
The current RTS proposal allows for a communicator to contain failed processes and continue to be used for all operations, including collectives, after acknowledging them. This matches closely to FT-MPI's BLANK mode. The user can use MPI_Comm_split() to get the equivalent of SHRINK if they need it.<br>
<br>
The suggested modification allows for only/primarily a SHRINK-like mode in order to have full functionality in the communicator. As discussed on the previous call, one can get the BLANK mode by adding a library on top of MPI that virtualizes the communicators to create shadow communicators. The argument for the SHRINK mode is that it is -easier- to pass/explain.<br>
<br>
The reason we chose BLANK was derived from the literature reviewed, code examples available, and feedback from application groups. >From which there seemed to be a strong demand for the BLANK mode. In fact, I had a difficult time finding good use cases for the SHIRNK mode (I'm still looking though). Additionally, a BLANK mode seems also to make it easier to reason about process recovery. To reason about process recovery (something like FT-MPI's REBUILD mode) one needs to be able to reason about the missing processes without changing the identities of the existing processes, which can be difficult in a SHRINK mode. So from this review it seemed that there was an application demand for a BLANK-like mode for the RTS proposal.<br>
<br>
In light of this background, it is concerning to me to advise these application users that MPI will not provide the functionality they require, but they have to depend upon a non-standard, third-party library because we shied away from doing the right thing by them. This background is advised from my review of the state of the art, but others may have alternative evidence/commentary to present as well that could sway the discussion. It just seems like a weak argument that we should do the easy thing at the expense of doing the right thing by the application community.<br>
<br>
<br>
I certainly meant this email to stimulate conversation for the teleconference tomorrow. In particular, I would like those on the list with experience building ABFT/Natural FT applications/libraries (UTK?) to express their perspective on this topic. Hopefully they can help guide us towards the right solution, which might just be a SHRINK-like mode.<br>
<br>
<br>
-- Josh<br>
<br>
--<br>
Joshua Hursey<br>
Postdoctoral Research Associate<br>
Oak Ridge National Laboratory<br>
<a href="http://users.nccs.gov/~jjhursey" target="_blank">http://users.nccs.gov/~jjhursey</a><br>
</div>_______________________________________________<br>
mpi3-ft mailing list<br>
<a href="mailto:mpi3-ft@lists.mpi-forum.org">mpi3-ft@lists.mpi-forum.org</a><mailto:<a href="mailto:mpi3-ft@lists.mpi-forum.org">mpi3-ft@lists.mpi-forum.org</a>><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft</a><br>
<br>
<br>
_______________________________________________<br>
mpi3-ft mailing list<br>
<a href="mailto:mpi3-ft@lists.mpi-forum.org">mpi3-ft@lists.mpi-forum.org</a><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft</a><br>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>Joshua Hursey<br>Postdoctoral Research Associate<br>Oak Ridge National Laboratory<br><a href="http://users.nccs.gov/~jjhursey" target="_blank">http://users.nccs.gov/~jjhursey</a><br>
</div>
_______________________________________________<br>mpi3-ft mailing list<br><a href="mailto:mpi3-ft@lists.mpi-forum.org">mpi3-ft@lists.mpi-forum.org</a><br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft</blockquote></div><br></div></div></div></body></html>