[Mpi3-rma] RMA proposal 1 update

Fri May 21 18:43:16 CDT 2010

>For ARMCI, I'm pretty sure I would do
some sort of allreduce-and-iterate, yes. We could maybe do some tricks
to minimize how many times we have to >iterate, but I haven't thought about
this much yet. And, yes, we can easily separate 1-sided vs 2-sided completion
counting requirements in ARMCI. In >MPI, it would still be fairly trivial;
there is enough "glue" from 1-sided or 2-sided into DCMF that
we could do the completion counting only on 1-sided. 

May be you dont need allreduce+iterate (or may be I am misunderstanding this)...
consider an allflushall exchange on 8 processors (first row is process rank and 36 is the pending count process 3 has for process 6):
1                                2                                3                                4                                5                                6                                  7                                811 12 13 14 15 16 17 18 21 22 23 24 25 26 27 28 31 32 33 34 35 36 37 38 41 42 43 44 45 46 47 48 51 52 53 54 55 56 57 58 61 62 63 64 65 66 67 68 71 72 73 74 75 76 77 78 81 82 83 84 85 86 87 88Step 1 has each process exchanging 4 elements and syncing 1 values. After step 1:11 12 13 14 51 52 53 54 21 22 23 24 61 62 63 64 31 32 33 34 71 72 73 74 41 42 43 44 81 82 83 84 15 16 17 18 55 56 57 58 25 26 27 28 65 66 67 68 35 36 37 38 75 76 77 78 45 46 47 48 85 86 87 88Step 2 has each process exchanging 4 elements and syncing 2 values. After step 2:11 12 31 32 51 52 71 72 ......Step 3 has each process exchanging 4 elements and syncing 3 values
ARMCI_Barrier is allfence+MPI_Barrier. 
Sorry if this disrupts the flow of the discussion, I see the discussion has now touching on allflushall fitting with existing semantics.

From:
"Underwood, Keith D" <keith.d.underwood at intel.com>

To:
"MPI 3.0 Remote Memory Access working
group" <mpi3-rma at lists.mpi-forum.org>

Date:
05/20/2010 04:57 PM

Subject:
Re: [Mpi3-rma] RMA proposal 1 update

Sent by:
mpi3-rma-bounces at lists.mpi-forum.org

My point was, the way Jeff is doing
synchronization in NWChem is via a fenceall(); barrier(); on the equivalent
of MPI_COMM_WORLD. If I knew he was going to be primarily doing this (ie,
that he wanted to know that all nodes were synched), I would do something
like maintain counts of sent and received messages on each node. I could
then do something like an allreduce of those 2 ints over the tree to determine
if everyone is synched. There are probably some technical details that
would have to be worked out to ensure this works but it seems good from
10000 feet. 

Right now we do numprocs 0-byte get operations to make sure the torus is
flushed on each node. A torus operation is ~3us on a 512-way. It grows
slowly with number of midplanes. I'm sure a 72 rack longest Manhattan distance
noncongested pingpong is <10us, but I don't have the data in front of
me. 

Based on Doug’s email,
I had assumed you would know who you have sent messages to…  If you
knew that in a given fence interval the node had only sent distinct messages
to 1K other cores, you would only have 1K gets to issue.  Suck? 
Yes.  Worse than the tree messages?  Maybe, maybe not. 
There is definitely a cross-over between 1 and np outstanding messages
between fences where on the 1 side of things the tree messages are worse
and on the np side of things the tree messages are better.  There
is another spectrum based on request size where getting a response for
every request becomes an inconsequential overhead.  I would have to
know the cost of processing a message, the size of a response, and the
cost of generating that response to create a proper graph of that. 

A tree int/sum is roughly 5us on a 512-way and grows similarly. I would
postulate that a 72 rack MPI allreduce int/sum is on the order of 10us.

So you generate np*np messages vs 1 tree message. Contention and all the
overhead of that many messages will be significantly worse than even several
tree messages.

Oh, wait, so, you would
sum all sent and sum all received and then check if they were equal? 
And then (presumably) iterate until the answer was yes?  Hrm. 
That is more interesting.  Can you easily separate one-sided and two
sided messages in your counting while maintaining the performance of one-sided
messages?

Doug’s earlier answer
implied you were going to allreduce a vector of counts (one per rank) and
that would have been ugly.   I am assuming you would do at least
2 tree messages in what I believe you are describing, so there is still
a crossover between n*np messages and m tree messages (where n is the number
of outstanding requests between fencealls and 2 <= m <= 10), and
the locality of communications impacts that crossover…  

BTW, can you actually
generate messages fast enough to cause contention with tiny messages?

Anytime I know that an operation is
collective, I can almost guarantee I can do it better than even a good
pt2pt algorithm if I am utilizing our collective network. I think on machines
that have remote completion notification an allfenceall() is just a barrier(),
and since fenceall(); barrier(); is going to be replaced by allfenceall(),
it doesn't seem to me like it is any extra overhead if allfenceall() is
just a barrier() for you. 

My concerns are twofold: 
1) we are talking about adding collective completion to passive target
when active target was the one designed to have collective completion. 
That is semantically and API-wise a bit ugly.  2) I think the allfenceall()
as a collective will optimize to the case where you have outstanding requests
to everybody and I believe that will be slower than the typical  case
of having outstanding requests to some people.  I think that users
would typically call allfenceall() rather than fenceall() + barrier() and
then they would see a performance paradox:  the fenceall() + barrier()
could be substantially faster when you have a “small” number of peers
you are communicating with in this iteration.  I am not at all worried
about the overhead of allfenceall() for networks with remote completion. 

Keith

From:

"Underwood, Keith D"
<keith.d.underwood at intel.com>

To:

"MPI 3.0 Remote Memory Access working
group" <mpi3-rma at lists.mpi-forum.org>

Date:

05/20/2010 09:19 AM

Subject:

Re: [Mpi3-rma] RMA proposal 1 update

Sent by:

mpi3-rma-bounces at lists.mpi-forum.org

> What is available in GA itself isn't really relevant to the Forum.
 We

> need the functionality that enables someone to implement GA

> ~~~efficiently~~~ on current and future platforms.  We know ARMCI
is

> ~~~necessary~~~ to implement GA efficiently on some platforms, but

> Vinod and I can provide very important cases where it is ~~~not

> sufficient~~~.

Then let's enumerate those and work on a solution.

> The reason I want allfenceall is because a GA sync requires every

> process to fence all remote targets.  This is combined with a
barrier,

> hence it might as well be a collective operation for everyone to fence

> all remote targets.  On BGP, implementing GA sync with fenceall
from

> every node is hideous compared to what I can imagine can be done with

> active-message collectives.  I would bet a kidney it is hideous
on

> Jaguar.  Vinod can sell my kidney in Singapore if I'm wrong.

> 

> The argument for allfenceall is the same as for sparse collectives.

> If there is an operation which could be done with multiple p2p calls,

> but has a collective character, it is guaranteed to be no worse to

> allow an MPI runtime to do it collectively.  I know that many

> applications will generate a sufficiently dense one-sided

> communication matrix to justify allfenceall.

So far, the argument I have heard for allflushall is:  BGP does not
give remote completion information to the source.  Surely making it
collective would be better. 

When I challenged that and asked for an implementation sketch, the implementation
sketch provided is demonstrably worse for many scenarios than calling flushall
and a barrier.  It would be a lot easier for the IBM people to do
the math to show where the crossover point is, but so far, they haven't.

> If you reject allfenceall, then I expect, and for intellectual

> consistency demand, that you vigorously protest against sparse

> collectives when they are proposed on the basis that they can

> obviously be done with p2p efficiently already.  Heck, why not
also

> deprecate all MPI_Bcast etc. since some on some networks it might
not

> be faster than p2p?

MPI_Bcast can ALWAYS be made faster than a naïve implementation over p2p.
 That is the point of a collective.  

Ask Torsten how much flak I gave him over some of the things he has proposed
for this reason.  Torsten made a rational argument for sparse collectives
that they convey information that the system can use successfully for optimization.
 I'm not 100% convinced, but he had to make that argument.  

> It is really annoying that you are such an obstructionist.  It
is

> extremely counter-productive to the Forum and I know of no one

I am attempting to hold all things to the standards set for MPI-3:

1) you need a use case.

2) you need an implementation

Now, I tend to think that means you need an implementation that helps your
use case.  In this particular case, you are asking to add collective
completion to a one-sided completion model.  This is fundamentally
inconsistent with the design of MPI RMA, which separates active target
(collective completion) from passive target (one-sided completion).  This
maps well to much of the known world of PGAS-like models:  CoArray
Fortran uses collective completion and UPC uses one-sided completion (admittedly,
a call to barrier will give collective completion in UPC, but that is because
a barrier without completion is meaningless).  This mixture of the
two models puts us at risk of always getting poor one-sided completion
implementations, since there is the "out" of telling people to
call the collective completion routine.  This would effectively gut
the advantages of passive target.  

So far, we have proposed adding:

1) Completion independent of synchronization

2) Some key remote operations

3) an ability to operate on the full window in one epoch

In my opinion, adding collective communication to passive target is a much
bigger deal.

> deriving intellectual benefit from the endless stream of protests
and

> demands for OpenSHMEM-like behavior.  As the ability to implement
GA

> on top of MPI-3 RMA is a stated goal of the working group, I feel
no

> shame in proposing function calls which are motivated entirely by
this

> purpose.

Endless stream of demands for OpenSHMEM-like behavior?  I have asked
(at times vigorously) for a memory model that would support the UPC memory
model.  The ability to support UPC is also in that stated goal along
with implementing GA.  I have used SHMEM as an example of that memory
model being done in an API and having hardware support from vendors.  I
have also argued that the memory model that supports UPC would be attractive
to SHMEM users and that OpenSHMEM is likely to be a competitor for mind
share for RMA-like programming models.  I have lost that argument
to the relatively vague "that might make performance worse in some
cases".  I find that frustrating, but I don't think I have raised
it since the last meeting.

Keith

_______________________________________________

mpi3-rma mailing list

mpi3-rma at lists.mpi-forum.org

http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma

_______________________________________________

mpi3-rma mailing list

mpi3-rma at lists.mpi-forum.org

http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-rma/attachments/20100521/39e97ce9/attachment-0001.html>