<html><body>
<p>I did not see this until Monday morning but here are some late comments.<br>
<br>
Dick<br>
<br>
How do you plan to give a reasonable range of stake holders a chance to understand the questions and provide their answers?<br>
<br>
ISVs or "closed source product" providers should be queried. They need to deal with their customers who are delivered an MPI 3<br>
<br>
Buyers of closed source products are stake holders in this issue. If I am using an MPI based product in my business, how do I feel about needing to buy and validate new software when my MPI provider says his new release breaks my production software?<br>
<br>
Providers of MPI implementations may want to think about the implications of an influential customer saying "Ship MPI 3 if you want to but to keep my business you must also keep shipping and supporting the MPI 2.2 that my apps still use.<br>
<br>
Typical number of tasks in an MPI job may not be very informative. The question should (also?) try to get at:<br>
<br>
1) What is the largest job you tend run today and consider important (vs something run but not critical to the mission / business)<br>
2) What number of tasks do you anticipate wanting to run (and consider important) within the next few years?<br>
<br>
I do not see anything clearly addressing the question of "How many cores. CPUs. PEs (whatever term) would you hope/expect to be able to apply to your largest problems?" Much of the hybrid discussion is about solving problems with so many PEs that MPI cannot scale to enough tasks. It seems likely to me that someone may want to use a million processing elements and that MPI will never scale well to a million tasks.<br>
<br>
I assume question with the check boxes is asking about "error handlers". It refers to "error handles"<br>
<a href="http://mpi-forum.questionpro.com/akira/TakeSurvey;jsessionid=dcaTO_iOlMbIqcQeW98ts"><u><font size="4" color="#0000FF">http://mpi-forum.questionpro.com/akira/TakeSurvey;jsessionid=dcaTO_iOlMbIqcQeW98ts</font></u></a><font size="4"> </font><br>
<br>
I do not know what this question means. <br>
<br>
<b><font face="Verdana">MPI one-sided communication performance is more important to me than supporting a rich remote memory access (RMA) feature set.</font></b><font size="4"> </font><br>
<br>
Having been involved in several email discussions and knowing what is being pressed, I can guess the desired answer. We cannot legitimately make decisions based on such an ambiguous question:<font size="4"> What expectations of typical MPI communication are included in "rich"? Is using any communicator except MPI_COMM_WORLD part of "rich"? Is getting a non-MPI_SUCCESS return code for an error part of rich? </font><br>
<br>
<font face="Verdana">"MPI application control of fault tolerance</font><font size="4">" </font>-- sounds too easy. Who would say no? I would be happy to have MPI_Tolerate_faults( yes | no)<br>
<br>
It would be helpful to be able to read the survey questions before starting to answer. I personally dislike answering a survey question without knowing what all the questions are. I must try to pick the best answer even though the question does not really fit my concern or I must ignore my concern on the assumption a better targeted question must be on the survey. (I cannot count the number of times I have started to answer a telephone survey only to say in the middle "We're done" because I have deduced the survey questions will not really help capture my views")<br>
<br>
<br>
<br>
<br>
Dick <br>
<br>
Dick Treumann - MPI Team <br>
IBM Systems & Technology Group<br>
Dept X2ZA / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601<br>
Tele (845) 433-7846 Fax (845) 433-8363<br>
<br>
<br>
<img width="16" height="16" src="cid:1__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt="Inactive hide details for Jeff Squyres ---11/14/2009 08:12:09 PM---Actually, when we put the questions in the survey web site, "><font color="#424282">Jeff Squyres ---11/14/2009 08:12:09 PM---Actually, when we put the questions in the survey web site, they came out slightly differently. H</font><br>
<br>
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">From:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">Jeff Squyres <jsquyres@cisco.com></font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">To:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">"Main MPI Forum mailing list" <mpi-forum@lists.mpi-forum.org></font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Date:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">11/14/2009 08:12 PM</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Subject:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">Re: [Mpi-forum] MPI user survey</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Sent by:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFCE3DFD71E9C8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">mpi-forum-bounces@lists.mpi-forum.org</font></td></tr>
</table>
<hr width="100%" size="2" align="left" noshade style="color:#8091A5; "><br>
<br>
<br>
<tt>Actually, when we put the questions in the survey web site, they came <br>
out slightly differently. Have a look here:<br>
<br>
</tt><tt><a href="http://mpi-forum.questionpro.com/">http://mpi-forum.questionpro.com/</a></tt><tt><br>
<br>
*** DO NOT GIVE THIS URL OUT TO USERS YET! ***<br>
<br>
Feel free to fill out the survey; we'll be clearing all the data on <br>
Monday evening so that it can "go live".<br>
<br>
<br>
<br>
<br>
On Nov 14, 2009, at 4:08 PM, Jeff Squyres (jsquyres) wrote:<br>
<br>
> Forum -- here's the questions that I took down on Friday morning.<br>
> Josh Hursey and I cleaned them up quite a bit, and we grabbed Bill<br>
> Gropp for 5 minutes on Saturday to give us a bit of spot feedback.<br>
> Here's the results.<br>
><br>
> *** Please send comments by Monday evening so that we can get these<br>
> posted on a web site. Thanks.<br>
><br>
> ------------------------<br>
> x. Which of the following best describes you?<br>
> - User of MPI applications<br>
> - MPI application developer<br>
> - MPI implementer<br>
> - Academic educator, student, or researcher<br>
> - Program / project management<br>
> - Other ________________<br>
><br>
> x. Rate your familiarity with the MPI standard?<br>
> - 1/not familiar at all ... 5/extremely familiar<br>
><br>
> x. Think of an MPI application that you run frequently. What is the<br>
> typical number of MPI processes per job that you run? (select all<br>
> that apply)<br>
> - 1-16 MPI processes<br>
> - 17-64 MPI processes<br>
> - 65-512 MPI processes<br>
> - 513-2048 MPI processes<br>
> - 2049 MPI processes or more<br>
> - I don't know<br>
><br>
> x. Using the same MPI application from #3, what is the typical number<br>
> of MPI processes that you run per node? (select all that apply)<br>
> - 1 MPI process<br>
> - 2-3 MPI processes<br>
> - 4-7 MPI processes<br>
> - 8-15 MPI processes<br>
> - 16 MPI processes or more<br>
> - I don't know<br>
><br>
> x. Using the same application from #3, is it a 32 or 64 bit <br>
> application?<br>
> (select all that apply)<br>
> - 32 bit<br>
> - 64 bit<br>
> - I don't know<br>
> - Other: _________________<br>
><br>
> x. Which of the following do your *any* of your MPI applications use?<br>
> (select all that apply)<br>
> - Threads<br>
> - OpenMP<br>
> - Shmem<br>
> - Global Arrays<br>
> - Co-processors / accelerators<br>
> - PGAS languages<br>
> - I don't know<br>
> - Other: ______________<br>
><br>
> x. How important are each of the following sets of MPI functionality<br>
> to *any* of your MPI applications?<br>
> 1/not important at all ... 5/very important<br>
> - Point-to-point communications<br>
> - Collective communications<br>
> - Derived / complex datatypes<br>
> - Communicators other than MPI_COMM_WORLD<br>
> - Graph or Cartesian process topologies<br>
> - Error handles / error checking<br>
> - Dynamic MPI processes (spawn, connect/accept, join)<br>
> - One-sided communication<br>
> - Generalized requests<br>
> - Parallel I/O<br>
> - "PMPI" profiling interface<br>
> - Multi-threaded applications (for example, MPI_THREAD_MULTIPLE)<br>
> - Other: ______________<br>
> If you marked any set with 1 or 2, please explain why.<br>
> __________<br>
><br>
> x. Are any of your MPI applications written to use the MPI C++<br>
> bindings?<br>
> - Yes<br>
> - No<br>
> - I don't know<br>
><br>
> x. I expect to be able to upgrade to an MPI-3 implementation and still<br>
> be able to run my legacy MPI aplications *without recompiling*.<br>
> Strongly agree/1 ...... Strongly disagree/5<br>
> Open comment: _________________________<br>
><br>
> x. I expect to be able to upgrade to an MPI-3 implementation and only<br>
> need to recompile my legacy MPI applications *with no source code<br>
> changes*.<br>
> Strongly agree/1 ....... Strongly disagree/5<br>
> Open comment: _________________________<br>
><br>
> x. My MPI application would benefit from being able to reference more<br>
> than 2^31 data items in a single MPI function invocation.<br>
> Strongly agree/1 ....... Strongly disagree/5<br>
> Open comment: _________________________<br>
><br>
> x. MPI one-sided communication performance is more important to me<br>
> than supporting a rich remote memory access (RMA) feature set.<br>
> Strongly agree/1 ....... Strongly disagree/5<br>
> Open comment: _________________________<br>
><br>
> x. The following are a list of topics that the MPI Forum is<br>
> considering for MPI-3. Rank them in order of importance to your<br>
> MPI applications:<br>
> - Non-blocking collective communications<br>
> - Revamped one-sided communications (compared to MPI-2.2)<br>
> - MPI application control of fault tolerance<br>
> - New Fortran bindings (type safety, etc.)<br>
> - "Hybrid" programming (MPI in conjunction with threads, <br>
> OpenMP, ..)<br>
> - Standardized third-party MPI tool support<br>
> - Other: ______________<br>
><br>
> x. What *ONE THING* would you like to see added or improved in the MPI<br>
> standard?<br>
> _____________<br>
><br>
> x. Rate the following in order of importance to your MPI applications:<br>
> - Performance<br>
> - Feature-rich API<br>
> - Run-time reliability<br>
> - Scalability to large numbers of MPI processes<br>
> - Integration with other communication protocols /<br>
><br>
> x. Did you attend the MPI Forum BOF at SC09?<br>
> - Yes<br>
> - No<br>
><br>
> x. Use the space below to provide any other information, suggestions,<br>
> or comments to the MPI Forum.<br>
> ________________________<br>
><br>
><br>
> --<br>
> Jeff Squyres<br>
> jsquyres@cisco.com<br>
><br>
> _______________________________________________<br>
> mpi-forum mailing list<br>
> mpi-forum@lists.mpi-forum.org<br>
> </tt><tt><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum</a></tt><tt><br>
><br>
<br>
<br>
-- <br>
Jeff Squyres<br>
jsquyres@cisco.com<br>
<br>
_______________________________________________<br>
mpi-forum mailing list<br>
mpi-forum@lists.mpi-forum.org<br>
</tt><tt><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum</a></tt><tt><br>
</tt><br>
<br>
</body></html>