May302012

MPI-3.0 is coming soon! Updates from the Japan Meeting.

HPC • MPI

mpi3v2

The Japan MPI Forum was rather “mild” until the last day where we had all the votes. Several controversial things came up for vote and many things that were not ready were pushed for a vote. We were 16 organizations eligible for voting and each ticket would only need 9 yes votes to get in, rather small imho.

While I am not 100% sure about the decision making process in the Forum, I think we made mostly sane decisions (some exception are of course strengthening this rule :-)).

Executive summary:

  • no fault tolerance for MPI-3.0: the Forum decided against the proposal
  • no “true” nonblocking I/O functions in MPI-3.0
  • no helper threads in MPI-3.0
  • removing the C++ bindings passed the first vote — scary!

Now to the detailed actions/votes:
First Votes

  • the small fixes #187 and #192 passed
  • #194 (allow non-rectangular MPI_Dims_create) was withdrawn based on comments
  • #195 (topology awareness in MPI_Dims_create) was rejected because the ticket was obviously not ready/clean
  • #217 (helper Threads) was rejected, it was always controversial
  • #256 (MPI_PROC_NULL behavior for MPI_PROBE) passed
  • #271 (functions to query MPI_Info object) passed, George raised an issue with the naming that we should fox before the final release
  • #273 (immediate versions of nonblocking collective I/O routines) was rejected, got very close but raised the concerns that there is no optimized implementation even for the split I/O right now
  • #278 (update examples to not use deprecated constructs) passed
  • #281 (remove C++ bindings) passed, unfortunately, long live C++ exceptions!
  • #294 (MPI_UNWEIGHTED should not be NULL) passed (clearly)
  • #300 (minor issue in One Sided) passed (clearly)
  • #303 (move MPI-2 deprecated functions to new “Removed interfaces”) was rejected after some discussion
  • #313 (fixing init init and finalize) passed
  • #310 (clarify MPI behavior when multiple MPI processes run in the same address space) was rejected because it was felt that it”s inconsistent with #313
  • #317 (correct error related to MPI_REQUEST_FREE) passed (trivial fix)
  • #323 (new FT proposal) was rejected, it was controversial before and seemed to be edited until the last second. #326 and #327 were withdrawn based on that vote. So this means essentially no FT for MPI-3.0
  • #328 (fix MPI_PROC_NULL behavior for mprobe/improbe/mrecv/imrecv) passed

Second Votes

  • the two simple (nearly ticket-0) RMA changes #308 and #309 passed unanimously
  • #284 (allocate shared memory window) passed – woohoo! I thought the idea was dead a long time ago!
  • #280 (hindexed_block) passed, as expected
  • #168 (nonblocking communicator duplication) passed
  • #272 (remove C++ bindings in nonblocking colls) passed 🙁
  • #286 (noncollective communicator creation) passed!
  • #305 (update MPI_Intercomm_create to use collective tag space) passed

May292012

Arrived in Japan and the Earth quaked immediately

Travel

In part two of my crazy travel (Champaign -> Zurich -> Lugano -> Zurich -> Tsukuba, Japan -> Zurich -> Champaign), I arrived in Japan. The flight was great, I love Turkish airlines! The bus was not so great (especially the very suboptimal schedule).

I am staying in Hotel Shinonome, where I stayed before (last time in 2008) and I like it very much. It’s somewhat hard to communicate with the staff (hand-wavy sign language helps a lot, but the biggest challenge was to ask for the wireless password — anyone knows sign language for that?? :-)).

During my first night, Tsukuba was hit by a 5.3 earthquake, which is not too bad but it was enough to wake me up and make my lamp fall from the table. When I first woke up I thought I was in the plane and turbulence started (I spent the night before in a rather turbulent flight :-)). Was an interesting experience, especially on the 9th floor!

earthquake

I’m now awaiting the last day of the MPI Forum and will then present my talk at Tokyotech and finally go on my way home on Friday! Hope the bus at 7:00am actually comes (they have a strange reservation system).

May252012

Advanced MPI-2.2 and MPI-3.0 Tutorial in Lugano

HPC • MPI • Travel

I gave my first tutorial about advanced MPI-2.2 usage and the upcoming MPI-3.0 standard in Lugano this week. Even though it was a lot of work to prepare the slides and especially the hands-on exercises, I have to say that it was a lot of fun. The user interaction was great and I learned a lot about how (Swiss) applications use MPI and what is required from future interfaces. The people at CSCS are also exceptionally nice and I very much enjoyed dinner with some!

The agenda and the slides of the tutorial are available on my tutorial website for this course. CSCS recorded everything and there may be a slide-cast soon.

The hands-on experience was also great! I was not expecting that so many folks would complete the exercises. I also realized that some things are somewhat non-intuitive in MPI. A good learning experience for me!

I’m looking forward to present more of those tutorials! If you missed the one in Lugano, I will present a slightly shorter version of the same tutorial with Martin Schulz at ISC’12.

My time in Lugano was also great, it’s just such a beautiful place:
916
lugano1

And the train ride back to Zurich was also wonderful. Some impressions:
train1
train1
train1
train1