February 2008 - Mpi3-subsetting - mailman3.cels.anl.gov

Re: [Mpi3-subsetting] Where is archive?
by unknown＠example.com 29 Feb '08

29 Feb '08

Hi, Our WG is so young that we have not put anything up yet. There must be quite a few emails in the archive by now, however, including minutes of yesterday's meeting. Summary slides capturing the state of discussion so far will follow next week. Best regards. Alexander ________________________________ From: mpi3-subsetting-bounces_at_[hidden] [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of Richard Graham Sent: Friday, February 29, 2008 4:08 PM To: Richard Treumann Cc: mpi3-subsetting_at_[hidden] Subject: Re: [Mpi3-subsetting] Where is archive? The mailing lists at uiuc are no longer active, and at this stage just forward mail to lists.mpi-forum.org . This too will be turned off in about 2 weeks. Each working group has wiki space for such things, some use if more than others. This wg just started its work yesterday, so very little has been done, and we are at the stage of trying to define what we mean by subsetting. The wiki pages can be accessed from the meetings web page, meetings.mpi-forum.org, by following the MPI 3.0 link, and then going to what ever working group you are interested in. I have not looked at the subsetting wiki site, to see if anything has been put up on it yet. Rich On 2/29/08 9:27 AM, "Richard Treumann" <treumann_at_[hidden]> wrote: FYI - the mailing list web page: http://lists.cs.uiuc.edu/mailman/listinfo has links to most or all of the email lists I know of except this one. Is there an archive? Also - is there an overview proposal somewhere? Thanks Dick Treumann - MPI Team/TCEM IBM Systems & Technology Group Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601 Tele (845) 433-7846 Fax (845) 433-8363 --------------------------------------------------------------------- Intel GmbH Dornacher Strasse 1 85622 Feldkirchen/Muenchen Germany Sitz der Gesellschaft: Feldkirchen bei Muenchen Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer Registergericht: Muenchen HRB 47456 Ust.-IdNr. VAT Registration No.: DE129385895 Citibank Frankfurt (BLZ 502 109 00) 600119052 This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. *

1 0

Re: [Mpi3-subsetting] Where is archive?
by unknown＠example.com 29 Feb '08

29 Feb '08

The mailing lists at uiuc are no longer active, and at this stage just forward mail to lists.mpi-forum.org . This too will be turned off in about 2 weeks. Each working group has wiki space for such things, some use if more than others. This wg just started its work yesterday, so very little has been done, and we are at the stage of trying to define what we mean by subsetting. The wiki pages can be accessed from the meetings web page, meetings.mpi-forum.org, by following the MPI 3.0 link, and then going to what ever working group you are interested in. I have not looked at the subsetting wiki site, to see if anything has been put up on it yet. Rich On 2/29/08 9:27 AM, "Richard Treumann" <treumann_at_[hidden]> wrote: > FYI - the mailing list web page: http://lists.cs.uiuc.edu/mailman/listinfo has > links to most or all of the email lists I know of except this one. > > Is there an archive? > > Also - is there an overview proposal somewhere? > > Thanks > > Dick Treumann - MPI Team/TCEM > IBM Systems & Technology Group > Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601 > Tele (845) 433-7846 Fax (845) 433-8363 > *

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

Bronis, for the record: I do *not* advocate to get rid of datatypes! I think datatypes are a great thing for some parallel applications and they certainly should be used as a high-level abstraction. I've implemented scatter/gather list-based optimizations for modern NICs (IB). But on the other hand, there are many codes out there that do just not use datatypes. Codes that are only supposed to run in heterogeneous environments. Codes that use sockets instead of MPI. If we want to aim at this market, we need to simplify here. A simplification could be to use MPI_BYTE by default ;) but it would be better to get rid of the code and control-path overhead. Just to clarify my opinion, Torsten -- bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ ----- Indiana University | http://www.indiana.edu Open Systems Lab | http://osl.iu.edu/ 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA Lindley Hall Room 135 | +01 (812) 855-3608

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

Hi, > Thanks. As soon as there's a couple of non-blocking recvs out there, > waiting for them in reverse order requires tracking of the moment when > the receives were posted. In some cases this leads to extra fields and > data exchanges. How's that different from multiple Recvs in multiple threads. Hmm, I guess the threaded case is just undefined and the implementation is allowed to ignore ordering, right? > The footprint argument generally says that the library will be smaller. > This may be a minor matter for general purpose computers, but as soon as > you go to Petascale, you need every byte on the compute nodes for user > data, especially if dynamic libraries are not supported. ack > As for the collectives, many are implemented using SendRecv, and that > blocking call in turn often uses non-blocking communication. Classic > Alltoallv algorithm uses nonblocking calls, too. So, I'm not sure that > even unoptimized blocking collectives will always use only blocking > pt2pt. Yes, of course - again, the proposed "interface" is more of a logical nature. You can implement all blocking collectives with blocking p2p (it'll be slower). But ok, pulling all p2p functions in this interface is also unproblematic. So I don't really have a srong opinion here. Best, Torsten -- bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ ----- Indiana University | http://www.indiana.edu Open Systems Lab | http://osl.iu.edu/ 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA Lindley Hall Room 135 | +01 (812) 855-3608

1 0

Re: [Mpi3-subsetting] Some "stupid user" questions, comments.
by unknown＠example.com 29 Feb '08

29 Feb '08

Dear RIchard, Thanks. The more complicated the standard gets, the happier are the implementors. However, now we try to think like MPI users for a change, so, thanks for providing a reality check. Now, to one of your questions. An MPI_ANY_SOURCE MPI_Recv in multifabric environment means that a receive has to be posted somehow to more than one fabric in the MPI device layer. Once one of them gets the message, the posted receives should be cancelled on other fabrics. Now, what if they've already matched and started to receive something? What if they cannot cancel a posted receive? And so on. There are 3 to 5 ways to deal with this situation, with and without actually posting a receive, but none of them is good enough if you ask me. That's why there are 3 to 5 of them, actually. And all of them complicate the progress engine - the heart of an MPI implementation - at exactly the spot where one wants things simple and fast. This means that most of the time we fight these repercussions and curse the MPI_ANY_SOURCE. Or, looping back to the beginning of this message, we actually never stop blessing MPI_ANY_SOURCE. Fighting this kind of trouble is what we are paid for. ;) Best regards. Alexander ________________________________ From: mpi3-subsetting-bounces_at_[hidden] [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of Richard Barrett Sent: Friday, February 29, 2008 2:50 PM To: mpi3-subsetting_at_[hidden] Subject: [Mpi3-subsetting] Some "stupid user" questions, comments. Hi folks, I'm still sorting things out in my mind, so perhaps this note is just me talking to myself. But should you feel so compelled to sort through it, I would appreciate any feedback you might offer; and it will make me a more informed participant. I see two main perspectives: the user and the implementer. I come from the user side, so I feel comfortable in positing that user confusion over the size of the standard is really a function of presentation. That is, most of us get our information regarding using MPI directly from the standard. For me, this is the _only_ standard I've ever actually read! Perhaps I am missing out on thousands of C and Fortran capabilities, but sometimes ignorance is bliss. That speaks highly to the MPI specification presentation; however it need not be the case. An easy solution to the "too many routines" complaint is a tutorial/book/chapter on the basics, with pointers to further information. And in fact these books exist. That said, I hope that MPI-3 deprecates a meaningful volume of functionality. >From the implementer perspective, there appear to be two goals. First is to ease the burden with regard to the amount of functionality that must be supported. (And we users don't want to hear of your whining, esp. from a company the size of Intel :) Second, which overlaps with user concerns, is performance. That is, by defining a small subset of functionality, strong performance (in some sense, e.g. speed or memory requirements) can be realized. At the risk of starting too detailed a discussion at this early point (as well as exposing my ignorance:), I will throw out a few situations for discussion. 1. What would such a subset would imply with regard to what I view as support functionality, such as user-defined datatypes, topologies, etc? Ie could this support be easily provided, say by cutting-and-pasting from the full implementation you will still provide? (I now see Torsten recommends excluding datatypes, but what of other stuff?) 2. Even more broadly (and perhaps very ignorantly), can I simply link in both libraries, like -lmpi_subset -lmpi, getting the good stuff from the former and the excluded functionality from the latter? In addition to the application developers use of MPI, all large application programs I've dealt with make some use of externally produced libraries (a "very good thing" imo), which probably exceed the functionality in a "subset" implementation. 3. I (basically) understand the adverse performance effects of allowing promiscuous receives (MPI_ANY_SOURCE). However, this is a powerful capability for many codes, and used only in moderation, eg for setting up communication requirements (such as communication partners in unstructured, semi-structured, and dynamic mesh computations). In this case the sender knows its partner, but the receiver does not. A reduction(sum) is used to let each process know the number of communication partners from which it will receive data, the process posts that many promiscuous receives, which when satisfied lets it from then on specify the sender. So would it be possible to include this capability in a separate function, say the blocking send/recv, but not allow it in the non-blocking version? 4. Collectives: I can't name a code I've ever worked with that doesn't require MPI_Allreduce (though I wouldn't be surprised to hear of many), and this in a broad set of science areas. MPI_Bcast is also often used (but quite often only in the setup phase). I see MPI_Reduce used most often to collect timing information, so MPI_Allreduce would probably be fine as well. MPI_Gather is often quite useful, as is MPI_Scatter, but again often in setup. (Though often "setup" occurs once per time step.) Non-constant size versions are often used. And others can also no doubt offer strong opinions regarding inclusion of exclusion. But from an implementation perspective, what are the issues? In particular, is the basic infrastructure for these (and other collective operations) the same? A driving premise for supporting collectives is that the sort of performance driven capability under discussion is most needed by applications running at very large scale, which is where even very good collect implementations run into problems. 5. Language bindings and perhaps other things: With the expectation/hope that full implementations continue to be available, I could use them for code development, thus making use of things like type checking, etc. And does this latter use then imply the need for "stubs" for things like the (vaporous) Fortran bindings module, communicators (if only MPI_COMM_WORLD is supported), etc.? And presuming the answer to #2 is "no", could/should the full implementation "warn" me (preferably at compile time) when I'm using functionality that rules out use of the subset? 6. Will the profile layer still be supported? Generating usage can still be quantified using a full implementation, but performance would not be (at least in this manner), which would rule out an apples-to-apples comparison between a full implementation and the subset version with its advertised superior performance. (Of course an overall runtime could be compared, which is the final word, but a more detailed analysis is often preferred.) 7. If blocking and non-blocking are required of the subset, aren't these blocking semantics? MPI_Send: MPI_Isend ( ..., &req ); MPI_Wait ( ..., &req ); ----- MPI_Recv: MPI_Irecv ( ..., &req ); MPI_Wait ( &req ); - And speaking of this, are there performance issues associated with variants of MPI_Wait, eg MPI_Waitany, MPI_Waitsome? Finally, I'll officially register my concern with what I see as an increasing complexity in this effort, esp wrt "multiple subsets". I don't intend this comment to suppress ideas, but to help keep the beating the drum for simplicity, which I see as a key goal of this effort. If you read this far, thanks! My apologies if some of these issues have been previously covered. And if I've simply exposed myself as ignorant, I feel confident is stating that I am not alone - these questions will persist from others. :) Richard -- Richard Barrett Future Technologies Group, Computer Science and Mathematics Division, and Scientific Computing Group, National Center for Computational Science Oak Ridge National Laboratory http://ft.ornl.gov/~rbarrett On 2/28/08 1:04 PM, "mpi3-subsetting-request_at_[hidden]" <mpi3-subsetting-request_at_[hidden]> wrote: > Thank you for your time today. It was a very good discussion. Here's > what I captured (please add/modify what I may have missed): > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), Richard > Barrett (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > - Opens & introductions > > - Scope of the effort > - Rich > - Minimum subset consistent with the rest of MPI, for > performance/memory footprint optimization > - Danger of splitting MPI, hence against optional features in the > standard > - Both blocking & nonblocking belong to the core > - Torsten > - Some collectives may go into selectable subsets > - MPI_ANY_SOURCE considered harmful > - Leonid > - Flexible support for optional features, means for choosing and > advertising level of compliance/set of features > - See enclosed email for Alexander's POV > > - General discussion snapshots > - Support of subsets: some or all? If some, possible linkage problems > in static apps (or dead calls). If all, where's the gain? > - Optional: really optional (may be not present) or selectable (are > present but may be unused)? > - Performance penalty for unused subsets: implementation matter or > standard choice? > - Portability may be limited to certain class of applications (think > FT, master-slave runs) > - All we design needs to be implementable, complexity needs to be > controlled > - An ability to use certain set of subsets should not preclude pulling > in other modules if necessary > - Whatever we do, it should not conflict with the ABI efforts > - Need to stay nice and be nicer wrt to the libraries (think > threading) and keep things simple > - The simplification argument, if put first, may not be liked by some > > - Next steps > - Please comment on these minutes, and add/modify what I may have > missed > - I'll prepare a couple of slides by next week summarizing our > discussion so far; again, your feedback will be most welcome > - At the meeting, it may be great to meet F2F briefly and discuss any > eventual loose ends before the presentation at the Forum; I'll see to > this > > Best regards. > > Alexander > > -- > Dr Alexander Supalov > Intel GmbH > Hermuelheimer Strasse 8a > 50321 Bruehl, Germany > Phone: +49 2232 209034 > Mobile: +49 173 511 8735 > Fax: +49 2232 209029 > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. >

1 0

Some "stupid user" questions, comments.
by unknown＠example.com 29 Feb '08

29 Feb '08

Hi folks, I'm still sorting things out in my mind, so perhaps this note is just me talking to myself. But should you feel so compelled to sort through it, I would appreciate any feedback you might offer; and it will make me a more informed participant. I see two main perspectives: the user and the implementer. I come from the user side, so I feel comfortable in positing that user confusion over the size of the standard is really a function of presentation. That is, most of us get our information regarding using MPI directly from the standard. For me, this is the _only_ standard I've ever actually read! Perhaps I am missing out on thousands of C and Fortran capabilities, but sometimes ignorance is bliss. That speaks highly to the MPI specification presentation; however it need not be the case. An easy solution to the "too many routines" complaint is a tutorial/book/chapter on the basics, with pointers to further information. And in fact these books exist. That said, I hope that MPI-3 deprecates a meaningful volume of functionality. >From the implementer perspective, there appear to be two goals. First is to ease the burden with regard to the amount of functionality that must be supported. (And we users don't want to hear of your whining, esp. from a company the size of Intel :) Second, which overlaps with user concerns, is performance. That is, by defining a small subset of functionality, strong performance (in some sense, e.g. speed or memory requirements) can be realized. At the risk of starting too detailed a discussion at this early point (as well as exposing my ignorance:), I will throw out a few situations for discussion. 1. What would such a subset would imply with regard to what I view as support functionality, such as user-defined datatypes, topologies, etc? Ie could this support be easily provided, say by cutting-and-pasting from the full implementation you will still provide? (I now see Torsten recommends excluding datatypes, but what of other stuff?) 2. Even more broadly (and perhaps very ignorantly), can I simply link in both libraries, like -lmpi_subset -lmpi, getting the good stuff from the former and the excluded functionality from the latter? In addition to the application developers use of MPI, all large application programs I¹ve dealt with make some use of externally produced libraries (a ³very good thing² imo), which probably exceed the functionality in a ³subset² implementation. 3. I (basically) understand the adverse performance effects of allowing promiscuous receives (MPI_ANY_SOURCE). However, this is a powerful capability for many codes, and used only in moderation, eg for setting up communication requirements (such as communication partners in unstructured, semi-structured, and dynamic mesh computations). In this case the sender knows its partner, but the receiver does not. A reduction(sum) is used to let each process know the number of communication partners from which it will receive data, the process posts that many promiscuous receives, which when satisfied lets it from then on specify the sender. So would it be possible to include this capability in a separate function, say the blocking send/recv, but not allow it in the non-blocking version? 4. Collectives: I can't name a code I've ever worked with that doesn't require MPI_Allreduce (though I wouldn¹t be surprised to hear of many), and this in a broad set of science areas. MPI_Bcast is also often used (but quite often only in the setup phase). I see MPI_Reduce used most often to collect timing information, so MPI_Allreduce would probably be fine as well. MPI_Gather is often quite useful, as is MPI_Scatter, but again often in setup. (Though often ³setup² occurs once per time step.) Non-constant size versions are often used. And others can also no doubt offer strong opinions regarding inclusion of exclusion. But from an implementation perspective, what are the issues? In particular, is the basic infrastructure for these (and other collective operations) the same? A driving premise for supporting collectives is that the sort of performance driven capability under discussion is most needed by applications running at very large scale, which is where even very good collect implementations run into problems. 5. Language bindings and perhaps other things: With the expectation/hope that full implementations continue to be available, I could use them for code development, thus making use of things like type checking, etc. And does this latter use then imply the need for "stubs" for things like the (vaporous) Fortran bindings module, communicators (if only MPI_COMM_WORLD is supported), etc.? And presuming the answer to #2 is ³no², could/should the full implementation ³warn² me (preferably at compile time) when I¹m using functionality that rules out use of the subset? 6. Will the profile layer still be supported? Generating usage can still be quantified using a full implementation, but performance would not be (at least in this manner), which would rule out an apples-to-apples comparison between a full implementation and the subset version with its advertised superior performance. (Of course an overall runtime could be compared, which is the final word, but a more detailed analysis is often preferred.) 7. If blocking and non-blocking are required of the subset, aren't these blocking semantics? MPI_Send: MPI_Isend ( ..., &req ); MPI_Wait ( ..., &req ); ----- MPI_Recv: MPI_Irecv ( ..., &req ); MPI_Wait ( &req ); - And speaking of this, are there performance issues associated with variants of MPI_Wait, eg MPI_Waitany, MPI_Waitsome? Finally, I¹ll officially register my concern with what I see as an increasing complexity in this effort, esp wrt ³multiple subsets². I don¹t intend this comment to suppress ideas, but to help keep the beating the drum for simplicity, which I see as a key goal of this effort. If you read this far, thanks! My apologies if some of these issues have been previously covered. And if I've simply exposed myself as ignorant, I feel confident is stating that I am not alone - these questions will persist from others. :) Richard -- Richard Barrett Future Technologies Group, Computer Science and Mathematics Division, and Scientific Computing Group, National Center for Computational Science Oak Ridge National Laboratory http://ft.ornl.gov/~rbarrett On 2/28/08 1:04 PM, "mpi3-subsetting-request_at_[hidden]" <mpi3-subsetting-request_at_[hidden]> wrote: > Thank you for your time today. It was a very good discussion. Here's > what I captured (please add/modify what I may have missed): > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), Richard > Barrett (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > - Opens & introductions > > - Scope of the effort > - Rich > - Minimum subset consistent with the rest of MPI, for > performance/memory footprint optimization > - Danger of splitting MPI, hence against optional features in the > standard > - Both blocking & nonblocking belong to the core > - Torsten > - Some collectives may go into selectable subsets > - MPI_ANY_SOURCE considered harmful > - Leonid > - Flexible support for optional features, means for choosing and > advertising level of compliance/set of features > - See enclosed email for Alexander's POV > > - General discussion snapshots > - Support of subsets: some or all? If some, possible linkage problems > in static apps (or dead calls). If all, where's the gain? > - Optional: really optional (may be not present) or selectable (are > present but may be unused)? > - Performance penalty for unused subsets: implementation matter or > standard choice? > - Portability may be limited to certain class of applications (think > FT, master-slave runs) > - All we design needs to be implementable, complexity needs to be > controlled > - An ability to use certain set of subsets should not preclude pulling > in other modules if necessary > - Whatever we do, it should not conflict with the ABI efforts > - Need to stay nice and be nicer wrt to the libraries (think > threading) and keep things simple > - The simplification argument, if put first, may not be liked by some > > - Next steps > - Please comment on these minutes, and add/modify what I may have > missed > - I'll prepare a couple of slides by next week summarizing our > discussion so far; again, your feedback will be most welcome > - At the meeting, it may be great to meet F2F briefly and discuss any > eventual loose ends before the presentation at the Forum; I'll see to > this > > Best regards. > > Alexander > > -- > Dr Alexander Supalov > Intel GmbH > Hermuelheimer Strasse 8a > 50321 Bruehl, Germany > Phone: +49 2232 209034 > Mobile: +49 173 511 8735 > Fax: +49 2232 209029 > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. >

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

OK, thanks. -----Original Message----- From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] Sent: Friday, February 29, 2008 2:17 PM To: Supalov, Alexander Cc: mpi3-subsetting_at_[hidden] Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09 Alexander: It is the vast majority of scientific applications. It is not just ones that need sparse data structures. A stencil application that uses dense matrices has strided non-contiguous data transfers for half (2D) or more (3D or more complex stencils) of its communication. Non-contiguous communication is the reality of distributed memory computing... I am fine with letting this rest but my point is that an emphasis on performance by implementers should be the case for datatypes... Bronis On Fri, 29 Feb 2008, Supalov, Alexander wrote: > Dear Bronis, > > Thanks. What scientific computing codes do you mean here - chemistry, > structural mechanics, fluid dynamics, genomics, something else? Or do > you speak generally of any code that needs sparse data structures? If > so, what's your estimate of the relative number of such codes compared > to those that do not need sparse datatypes? In what domain? > > The right doze of vendor motivation not to do wrong things is a good > point, I'll consider it. > > Finally, the constant stride copying was but an example when inlining > may help to users achieve higher performance. There may be other > examples known in the scientific computing area. However, since > performance is not primary goal for datatypes, I suggest we let this > matter rest for a while. > > Best regards. > > Alexander > > -----Original Message----- > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > Sent: Friday, February 29, 2008 10:11 AM > To: Supalov, Alexander > Cc: mpi3-subsetting_at_[hidden] > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > ww09 > > > Alexander: > > Re: > > Thanks. I understand your motivation. When you say "most real > > applications" - what applications do you mean? At least, in what area? > > ? Scientific computing... > > > For the NIC part, the stress was on "here". In my opinion, subsetting > is > > not about making things more complicated, more challenging to the > > implementors, or to the underlying hardware. It's about making things > > simple, easy to use, and easy to implement - including implementation > of > > only those features your users actually need. That the implementation > > may be faster due to this is an added bonus, not the primary goal. > > The emphasis here should not be on creating a disincentive > for vendors to do the right thing... > > > Still, regarding user side copying. Yes, when people do this one > wonders > > why. There's a reason, apart from them: 1) not caring about datatypes > > and their complexity and 2) not trusting their performance. A modern > > compiler can rather well optimize a loop with a constant stride, and > may > > have difficulty with an unknown stride. This is why explicit loops are > > sometimes indeed faster (much faster) in the resulting code than any > > generic implementation. > > Huh? What makes you think the user copying code is > in terms of constant stride? Generally, it varies > with the input. We are not talking about a simple > situation to optimize at the user level... > > Bronis > > > > > > Best regards. > > > > Alexander > > > > -----Original Message----- > > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > > Sent: Friday, February 29, 2008 6:20 AM > > To: Supalov, Alexander > > Cc: mpi3-subsetting_at_[hidden] > > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > ww09 > > > > > > Alexander: > > > > Most real applications need to send non-contiguous > > data. If they do not use datatypes then they are > > doing the equivalent of either the packing/unpacking > > or smaller messages at the user level. This s hould > > be discouraged, not encouraged. A small savings > > in library object size is not ample reason to go > > against that. And, yes, we are after encouraging > > hardware vendors to provide the right hardware. > > > > Bronis > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > Hi, > > > > > > Thanks. I think the main thrust here is the library footprint (no > > > pack/unpack, etc.) and complexity of the user side of the datatype > > > interface, rather than performance. Many applications just don't > need > > > any of this, and never will. Why not translating this application > > > non-requirement into a minimum MPI subset? Same with > > communicator/group > > > management, etc. > > > > > > Moreover, homogeneous installations that dominate HPC now don't > > actually > > > need any datatype support at all. They send chunks of bytes. This > may > > > change in the future, though. > > > > > > A minor performance implication is that without holes that are only > > > possible with derived datatypes, one does not need to track this, > > split > > > the critical path, and make special provisions inside the MPI device > > > layer to handle iov or such. > > > > > > The NIC capability argument is interesting, but it turns the > > discussion > > > on its head: we're not after motivating network vendors to provide > > > scatter/gather in hardware here, are we? Please clarify. > > > > > > Best regards. > > > > > > Alexander > > > > > > -----Original Message----- > > > From: mpi3-subsetting-bounces_at_[hidden] > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > Bronis > > > R. de Supinski > > > Sent: Friday, February 29, 2008 5:53 AM > > > To: mpi3-subsetting_at_[hidden] > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > > ww09 > > > > > > > > > All: > > > > > > OK, I have to respond to the notion that derived datatypes > > > limit performance. It is just not a reasonable position. > > > > > > Sure, if you can send contiguous locations, you will get > > > higher performance. The problem is that codes do not only > > > need to send contiguous data so that is not an adequate > > > reason to say derived datatypes limit performance. > > > > > > So, what is left? That there is some more efficient way > > > to send non-contiguous data? How? As multiple messages, > > > each of which send contiguous data? If so, then the > > > implementation could do this under the covers and the > > > datatypes are just a convenience for the user not to > > > have to specify the individual sends. OK, suppose that's > > > not the reason. Perhaps the user can do the copying into > > > a contiguous buffer and get better performance? While > > > I have seen this hold with some implementations, it is > > > absurd. There is no reason that I can discern as to why > > > the user should be able to deduce a better copying > > > mechanism than the MPI implementer. So, again, at worst, > > > the datatypes should be a convenience. Do you have an > > > alternative reason or a refutation of these opinions? > > > > > > What is more important, it is certainly possible to build > > > scatter/gather support into a NIC and achieve better > > > performance with datatypes than without. While there are > > > issues to be resolved for that (primarily the issue of > > > pinning memory), they are solvable with the right hardware > > > mechanism. Just because it does not yet exist is not > > > an adequate reason to say "Get rid of datatypes". OK, > > > you are not saying that but you are saying to deprecate > > > them in a sense. And saying you could send contiguous > > > sends more efficiently is a bad argument here. How do > > > datatypes cause inefficiency for that? How much is > > > that cost really? At what point do you hit where the > > > answer is "It would be faster not to compute anything"? > > > > > > Bronis > > > > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > > > Hi, > > > > > > > > Thanks. What subsets inside the current standard would you > propose? > > > What > > > > interfaces between them would you envision? > > > > > > > > Good idea about the optimization opportunities. Here's an initial > > > > combined list, with the main benefits as I see them. Please > > > > comment/extend. > > > > > > > > - Dynamic process support: less overhead in the progress engine, > > > easier > > > > global rank handling. > > > > - Heterogeneity: better memory footprint, easier data handling. > > > > - Derived datatypes (especially those with holes): better memory > > > > footprint. > > > > - MPI_ANY_SOURCE: faster, more simple multifabric progress. > > > > - File I/O: smaller requests, easier wait/test functions. > > > > - One-sided ops: no passive target w/o MPI calls - no extra > progress > > > > thread. > > > > - Communicator & group management: better memory footprint. > > > > - Message tagging: better support for stable dataflow exchanges, > > > smaller > > > > packets. > > > > - Non-blocking communication: easier ordering, simplified request > > > > handling. > > > > > > > > Best regards. > > > > > > > > Alexander > > > > > > > > -----Original Message----- > > > > From: mpi3-subsetting-bounces_at_[hidden] > > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > > > Torsten Hoefler > > > > Sent: Friday, February 29, 2008 5:08 AM > > > > To: mpi3-subsetting_at_[hidden] > > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff > telecon > > > > ww09 > > > > > > > > Hi, > > > > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), > > Richard > > > > Barrett > > > > > (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > > > just for the record, it's "IU" not "ISU" :-) > > > > > > > > > - Scope of the effort > > > > > - Rich > > > > > - Minimum subset consistent with the rest of MPI, for > > > > > performance/memory footprint optimization > > > > > - Danger of splitting MPI, hence against optional > features > > in > > > > the > > > > > standard > > > > I back that (danger of optional features for portability). I'd > > propose > > > > to split the current standard into mostly self-contained subsets > > that > > > > have clearly defined interfaces to the rest of the standard. Note: > > > this > > > > only defines logical interfaces, that does *not* define how those > > > things > > > > are to be implemented. This makes it easier to understand the > > standard > > > > and have separate (portable) libraries for the subsets, it does > not > > > > influence optimization possibilities by implementing everything in > a > > > > monolithic block (i.e., central progress). > > > > > > > > > - Both blocking & nonblocking belong to the core > > > > > - Torsten > > > > > - Some collectives may go into selectable subsets > > > > I see three subsets: blocking colls, non-blocking colls and > > > topological > > > > colls (maybe also blocking / non-blocking). > > > > > > > > > - MPI_ANY_SOURCE considered harmful > > > > I'd like to add datatypes and heterogeneity to this list (with > > regards > > > > to performance). Alexander mentioned the dynamics. I think we > should > > > > have a lit of items ready that could influence optimization > > > > possibilities significanty if they were to be announced by the > user > > > > before he can use them. That would give another strong argument > for > > > the > > > > subsetting. > > > > > > > > Best, > > > > Torsten > > > > > > > > -- > > > > bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ > > ----- > > > > Indiana University | http://www.indiana.edu > > > > Open Systems Lab | http://osl.iu.edu/ > > > > 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA > > > > Lindley Hall Room 135 | +01 (812) 855-3608 > > > > _______________________________________________ > > > > Mpi3-subsetting mailing list > > > > Mpi3-subsetting_at_[hidden] > > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > > --------------------------------------------------------------------- > > > > Intel GmbH > > > > Dornacher Strasse 1 > > > > 85622 Feldkirchen/Muenchen Germany > > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes > Schwaderer > > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > > VAT Registration No.: DE129385895 > > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > > > This e-mail and any attachments may contain confidential material > > for > > > > the sole use of the intended recipient(s). Any review or > > distribution > > > > by others is strictly prohibited. If you are not the intended > > > > recipient, please contact the sender and delete all copies. > > > > > > > > > > > > _______________________________________________ > > > > Mpi3-subsetting mailing list > > > > Mpi3-subsetting_at_[hidden] > > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > --------------------------------------------------------------------- > > > Intel GmbH > > > Dornacher Strasse 1 > > > 85622 Feldkirchen/Muenchen Germany > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > VAT Registration No.: DE129385895 > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > This e-mail and any attachments may contain confidential material > for > > > the sole use of the intended recipient(s). Any review or > distribution > > > by others is strictly prohibited. If you are not the intended > > > recipient, please contact the sender and delete all copies. > > > > > > > > --------------------------------------------------------------------- > > Intel GmbH > > Dornacher Strasse 1 > > 85622 Feldkirchen/Muenchen Germany > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > VAT Registration No.: DE129385895 > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > This e-mail and any attachments may contain confidential material for > > the sole use of the intended recipient(s). Any review or distribution > > by others is strictly prohibited. If you are not the intended > > recipient, please contact the sender and delete all copies. > > > > > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > > --------------------------------------------------------------------- Intel GmbH Dornacher Strasse 1 85622 Feldkirchen/Muenchen Germany Sitz der Gesellschaft: Feldkirchen bei Muenchen Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer Registergericht: Muenchen HRB 47456 Ust.-IdNr. VAT Registration No.: DE129385895 Citibank Frankfurt (BLZ 502 109 00) 600119052 This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

Alexander: It is the vast majority of scientific applications. It is not just ones that need sparse data structures. A stencil application that uses dense matrices has strided non-contiguous data transfers for half (2D) or more (3D or more complex stencils) of its communication. Non-contiguous communication is the reality of distributed memory computing... I am fine with letting this rest but my point is that an emphasis on performance by implementers should be the case for datatypes... Bronis On Fri, 29 Feb 2008, Supalov, Alexander wrote: > Dear Bronis, > > Thanks. What scientific computing codes do you mean here - chemistry, > structural mechanics, fluid dynamics, genomics, something else? Or do > you speak generally of any code that needs sparse data structures? If > so, what's your estimate of the relative number of such codes compared > to those that do not need sparse datatypes? In what domain? > > The right doze of vendor motivation not to do wrong things is a good > point, I'll consider it. > > Finally, the constant stride copying was but an example when inlining > may help to users achieve higher performance. There may be other > examples known in the scientific computing area. However, since > performance is not primary goal for datatypes, I suggest we let this > matter rest for a while. > > Best regards. > > Alexander > > -----Original Message----- > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > Sent: Friday, February 29, 2008 10:11 AM > To: Supalov, Alexander > Cc: mpi3-subsetting_at_[hidden] > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > ww09 > > > Alexander: > > Re: > > Thanks. I understand your motivation. When you say "most real > > applications" - what applications do you mean? At least, in what area? > > ? Scientific computing... > > > For the NIC part, the stress was on "here". In my opinion, subsetting > is > > not about making things more complicated, more challenging to the > > implementors, or to the underlying hardware. It's about making things > > simple, easy to use, and easy to implement - including implementation > of > > only those features your users actually need. That the implementation > > may be faster due to this is an added bonus, not the primary goal. > > The emphasis here should not be on creating a disincentive > for vendors to do the right thing... > > > Still, regarding user side copying. Yes, when people do this one > wonders > > why. There's a reason, apart from them: 1) not caring about datatypes > > and their complexity and 2) not trusting their performance. A modern > > compiler can rather well optimize a loop with a constant stride, and > may > > have difficulty with an unknown stride. This is why explicit loops are > > sometimes indeed faster (much faster) in the resulting code than any > > generic implementation. > > Huh? What makes you think the user copying code is > in terms of constant stride? Generally, it varies > with the input. We are not talking about a simple > situation to optimize at the user level... > > Bronis > > > > > > Best regards. > > > > Alexander > > > > -----Original Message----- > > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > > Sent: Friday, February 29, 2008 6:20 AM > > To: Supalov, Alexander > > Cc: mpi3-subsetting_at_[hidden] > > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > ww09 > > > > > > Alexander: > > > > Most real applications need to send non-contiguous > > data. If they do not use datatypes then they are > > doing the equivalent of either the packing/unpacking > > or smaller messages at the user level. This s hould > > be discouraged, not encouraged. A small savings > > in library object size is not ample reason to go > > against that. And, yes, we are after encouraging > > hardware vendors to provide the right hardware. > > > > Bronis > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > Hi, > > > > > > Thanks. I think the main thrust here is the library footprint (no > > > pack/unpack, etc.) and complexity of the user side of the datatype > > > interface, rather than performance. Many applications just don't > need > > > any of this, and never will. Why not translating this application > > > non-requirement into a minimum MPI subset? Same with > > communicator/group > > > management, etc. > > > > > > Moreover, homogeneous installations that dominate HPC now don't > > actually > > > need any datatype support at all. They send chunks of bytes. This > may > > > change in the future, though. > > > > > > A minor performance implication is that without holes that are only > > > possible with derived datatypes, one does not need to track this, > > split > > > the critical path, and make special provisions inside the MPI device > > > layer to handle iov or such. > > > > > > The NIC capability argument is interesting, but it turns the > > discussion > > > on its head: we're not after motivating network vendors to provide > > > scatter/gather in hardware here, are we? Please clarify. > > > > > > Best regards. > > > > > > Alexander > > > > > > -----Original Message----- > > > From: mpi3-subsetting-bounces_at_[hidden] > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > Bronis > > > R. de Supinski > > > Sent: Friday, February 29, 2008 5:53 AM > > > To: mpi3-subsetting_at_[hidden] > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > > ww09 > > > > > > > > > All: > > > > > > OK, I have to respond to the notion that derived datatypes > > > limit performance. It is just not a reasonable position. > > > > > > Sure, if you can send contiguous locations, you will get > > > higher performance. The problem is that codes do not only > > > need to send contiguous data so that is not an adequate > > > reason to say derived datatypes limit performance. > > > > > > So, what is left? That there is some more efficient way > > > to send non-contiguous data? How? As multiple messages, > > > each of which send contiguous data? If so, then the > > > implementation could do this under the covers and the > > > datatypes are just a convenience for the user not to > > > have to specify the individual sends. OK, suppose that's > > > not the reason. Perhaps the user can do the copying into > > > a contiguous buffer and get better performance? While > > > I have seen this hold with some implementations, it is > > > absurd. There is no reason that I can discern as to why > > > the user should be able to deduce a better copying > > > mechanism than the MPI implementer. So, again, at worst, > > > the datatypes should be a convenience. Do you have an > > > alternative reason or a refutation of these opinions? > > > > > > What is more important, it is certainly possible to build > > > scatter/gather support into a NIC and achieve better > > > performance with datatypes than without. While there are > > > issues to be resolved for that (primarily the issue of > > > pinning memory), they are solvable with the right hardware > > > mechanism. Just because it does not yet exist is not > > > an adequate reason to say "Get rid of datatypes". OK, > > > you are not saying that but you are saying to deprecate > > > them in a sense. And saying you could send contiguous > > > sends more efficiently is a bad argument here. How do > > > datatypes cause inefficiency for that? How much is > > > that cost really? At what point do you hit where the > > > answer is "It would be faster not to compute anything"? > > > > > > Bronis > > > > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > > > Hi, > > > > > > > > Thanks. What subsets inside the current standard would you > propose? > > > What > > > > interfaces between them would you envision? > > > > > > > > Good idea about the optimization opportunities. Here's an initial > > > > combined list, with the main benefits as I see them. Please > > > > comment/extend. > > > > > > > > - Dynamic process support: less overhead in the progress engine, > > > easier > > > > global rank handling. > > > > - Heterogeneity: better memory footprint, easier data handling. > > > > - Derived datatypes (especially those with holes): better memory > > > > footprint. > > > > - MPI_ANY_SOURCE: faster, more simple multifabric progress. > > > > - File I/O: smaller requests, easier wait/test functions. > > > > - One-sided ops: no passive target w/o MPI calls - no extra > progress > > > > thread. > > > > - Communicator & group management: better memory footprint. > > > > - Message tagging: better support for stable dataflow exchanges, > > > smaller > > > > packets. > > > > - Non-blocking communication: easier ordering, simplified request > > > > handling. > > > > > > > > Best regards. > > > > > > > > Alexander > > > > > > > > -----Original Message----- > > > > From: mpi3-subsetting-bounces_at_[hidden] > > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > > > Torsten Hoefler > > > > Sent: Friday, February 29, 2008 5:08 AM > > > > To: mpi3-subsetting_at_[hidden] > > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff > telecon > > > > ww09 > > > > > > > > Hi, > > > > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), > > Richard > > > > Barrett > > > > > (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > > > just for the record, it's "IU" not "ISU" :-) > > > > > > > > > - Scope of the effort > > > > > - Rich > > > > > - Minimum subset consistent with the rest of MPI, for > > > > > performance/memory footprint optimization > > > > > - Danger of splitting MPI, hence against optional > features > > in > > > > the > > > > > standard > > > > I back that (danger of optional features for portability). I'd > > propose > > > > to split the current standard into mostly self-contained subsets > > that > > > > have clearly defined interfaces to the rest of the standard. Note: > > > this > > > > only defines logical interfaces, that does *not* define how those > > > things > > > > are to be implemented. This makes it easier to understand the > > standard > > > > and have separate (portable) libraries for the subsets, it does > not > > > > influence optimization possibilities by implementing everything in > a > > > > monolithic block (i.e., central progress). > > > > > > > > > - Both blocking & nonblocking belong to the core > > > > > - Torsten > > > > > - Some collectives may go into selectable subsets > > > > I see three subsets: blocking colls, non-blocking colls and > > > topological > > > > colls (maybe also blocking / non-blocking). > > > > > > > > > - MPI_ANY_SOURCE considered harmful > > > > I'd like to add datatypes and heterogeneity to this list (with > > regards > > > > to performance). Alexander mentioned the dynamics. I think we > should > > > > have a lit of items ready that could influence optimization > > > > possibilities significanty if they were to be announced by the > user > > > > before he can use them. That would give another strong argument > for > > > the > > > > subsetting. > > > > > > > > Best, > > > > Torsten > > > > > > > > -- > > > > bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ > > ----- > > > > Indiana University | http://www.indiana.edu > > > > Open Systems Lab | http://osl.iu.edu/ > > > > 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA > > > > Lindley Hall Room 135 | +01 (812) 855-3608 > > > > _______________________________________________ > > > > Mpi3-subsetting mailing list > > > > Mpi3-subsetting_at_[hidden] > > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > > --------------------------------------------------------------------- > > > > Intel GmbH > > > > Dornacher Strasse 1 > > > > 85622 Feldkirchen/Muenchen Germany > > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes > Schwaderer > > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > > VAT Registration No.: DE129385895 > > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > > > This e-mail and any attachments may contain confidential material > > for > > > > the sole use of the intended recipient(s). Any review or > > distribution > > > > by others is strictly prohibited. If you are not the intended > > > > recipient, please contact the sender and delete all copies. > > > > > > > > > > > > _______________________________________________ > > > > Mpi3-subsetting mailing list > > > > Mpi3-subsetting_at_[hidden] > > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > --------------------------------------------------------------------- > > > Intel GmbH > > > Dornacher Strasse 1 > > > 85622 Feldkirchen/Muenchen Germany > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > VAT Registration No.: DE129385895 > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > This e-mail and any attachments may contain confidential material > for > > > the sole use of the intended recipient(s). Any review or > distribution > > > by others is strictly prohibited. If you are not the intended > > > recipient, please contact the sender and delete all copies. > > > > > > > > --------------------------------------------------------------------- > > Intel GmbH > > Dornacher Strasse 1 > > 85622 Feldkirchen/Muenchen Germany > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > VAT Registration No.: DE129385895 > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > This e-mail and any attachments may contain confidential material for > > the sole use of the intended recipient(s). Any review or distribution > > by others is strictly prohibited. If you are not the intended > > recipient, please contact the sender and delete all copies. > > > > > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > >

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

Dear Bronis, Thanks. What scientific computing codes do you mean here - chemistry, structural mechanics, fluid dynamics, genomics, something else? Or do you speak generally of any code that needs sparse data structures? If so, what's your estimate of the relative number of such codes compared to those that do not need sparse datatypes? In what domain? The right doze of vendor motivation not to do wrong things is a good point, I'll consider it. Finally, the constant stride copying was but an example when inlining may help to users achieve higher performance. There may be other examples known in the scientific computing area. However, since performance is not primary goal for datatypes, I suggest we let this matter rest for a while. Best regards. Alexander -----Original Message----- From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] Sent: Friday, February 29, 2008 10:11 AM To: Supalov, Alexander Cc: mpi3-subsetting_at_[hidden] Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09 Alexander: Re: > Thanks. I understand your motivation. When you say "most real > applications" - what applications do you mean? At least, in what area? ? Scientific computing... > For the NIC part, the stress was on "here". In my opinion, subsetting is > not about making things more complicated, more challenging to the > implementors, or to the underlying hardware. It's about making things > simple, easy to use, and easy to implement - including implementation of > only those features your users actually need. That the implementation > may be faster due to this is an added bonus, not the primary goal. The emphasis here should not be on creating a disincentive for vendors to do the right thing... > Still, regarding user side copying. Yes, when people do this one wonders > why. There's a reason, apart from them: 1) not caring about datatypes > and their complexity and 2) not trusting their performance. A modern > compiler can rather well optimize a loop with a constant stride, and may > have difficulty with an unknown stride. This is why explicit loops are > sometimes indeed faster (much faster) in the resulting code than any > generic implementation. Huh? What makes you think the user copying code is in terms of constant stride? Generally, it varies with the input. We are not talking about a simple situation to optimize at the user level... Bronis > > Best regards. > > Alexander > > -----Original Message----- > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > Sent: Friday, February 29, 2008 6:20 AM > To: Supalov, Alexander > Cc: mpi3-subsetting_at_[hidden] > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > ww09 > > > Alexander: > > Most real applications need to send non-contiguous > data. If they do not use datatypes then they are > doing the equivalent of either the packing/unpacking > or smaller messages at the user level. This s hould > be discouraged, not encouraged. A small savings > in library object size is not ample reason to go > against that. And, yes, we are after encouraging > hardware vendors to provide the right hardware. > > Bronis > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > Hi, > > > > Thanks. I think the main thrust here is the library footprint (no > > pack/unpack, etc.) and complexity of the user side of the datatype > > interface, rather than performance. Many applications just don't need > > any of this, and never will. Why not translating this application > > non-requirement into a minimum MPI subset? Same with > communicator/group > > management, etc. > > > > Moreover, homogeneous installations that dominate HPC now don't > actually > > need any datatype support at all. They send chunks of bytes. This may > > change in the future, though. > > > > A minor performance implication is that without holes that are only > > possible with derived datatypes, one does not need to track this, > split > > the critical path, and make special provisions inside the MPI device > > layer to handle iov or such. > > > > The NIC capability argument is interesting, but it turns the > discussion > > on its head: we're not after motivating network vendors to provide > > scatter/gather in hardware here, are we? Please clarify. > > > > Best regards. > > > > Alexander > > > > -----Original Message----- > > From: mpi3-subsetting-bounces_at_[hidden] > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > Bronis > > R. de Supinski > > Sent: Friday, February 29, 2008 5:53 AM > > To: mpi3-subsetting_at_[hidden] > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > ww09 > > > > > > All: > > > > OK, I have to respond to the notion that derived datatypes > > limit performance. It is just not a reasonable position. > > > > Sure, if you can send contiguous locations, you will get > > higher performance. The problem is that codes do not only > > need to send contiguous data so that is not an adequate > > reason to say derived datatypes limit performance. > > > > So, what is left? That there is some more efficient way > > to send non-contiguous data? How? As multiple messages, > > each of which send contiguous data? If so, then the > > implementation could do this under the covers and the > > datatypes are just a convenience for the user not to > > have to specify the individual sends. OK, suppose that's > > not the reason. Perhaps the user can do the copying into > > a contiguous buffer and get better performance? While > > I have seen this hold with some implementations, it is > > absurd. There is no reason that I can discern as to why > > the user should be able to deduce a better copying > > mechanism than the MPI implementer. So, again, at worst, > > the datatypes should be a convenience. Do you have an > > alternative reason or a refutation of these opinions? > > > > What is more important, it is certainly possible to build > > scatter/gather support into a NIC and achieve better > > performance with datatypes than without. While there are > > issues to be resolved for that (primarily the issue of > > pinning memory), they are solvable with the right hardware > > mechanism. Just because it does not yet exist is not > > an adequate reason to say "Get rid of datatypes". OK, > > you are not saying that but you are saying to deprecate > > them in a sense. And saying you could send contiguous > > sends more efficiently is a bad argument here. How do > > datatypes cause inefficiency for that? How much is > > that cost really? At what point do you hit where the > > answer is "It would be faster not to compute anything"? > > > > Bronis > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > Hi, > > > > > > Thanks. What subsets inside the current standard would you propose? > > What > > > interfaces between them would you envision? > > > > > > Good idea about the optimization opportunities. Here's an initial > > > combined list, with the main benefits as I see them. Please > > > comment/extend. > > > > > > - Dynamic process support: less overhead in the progress engine, > > easier > > > global rank handling. > > > - Heterogeneity: better memory footprint, easier data handling. > > > - Derived datatypes (especially those with holes): better memory > > > footprint. > > > - MPI_ANY_SOURCE: faster, more simple multifabric progress. > > > - File I/O: smaller requests, easier wait/test functions. > > > - One-sided ops: no passive target w/o MPI calls - no extra progress > > > thread. > > > - Communicator & group management: better memory footprint. > > > - Message tagging: better support for stable dataflow exchanges, > > smaller > > > packets. > > > - Non-blocking communication: easier ordering, simplified request > > > handling. > > > > > > Best regards. > > > > > > Alexander > > > > > > -----Original Message----- > > > From: mpi3-subsetting-bounces_at_[hidden] > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > > Torsten Hoefler > > > Sent: Friday, February 29, 2008 5:08 AM > > > To: mpi3-subsetting_at_[hidden] > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > > ww09 > > > > > > Hi, > > > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), > Richard > > > Barrett > > > > (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > > just for the record, it's "IU" not "ISU" :-) > > > > > > > - Scope of the effort > > > > - Rich > > > > - Minimum subset consistent with the rest of MPI, for > > > > performance/memory footprint optimization > > > > - Danger of splitting MPI, hence against optional features > in > > > the > > > > standard > > > I back that (danger of optional features for portability). I'd > propose > > > to split the current standard into mostly self-contained subsets > that > > > have clearly defined interfaces to the rest of the standard. Note: > > this > > > only defines logical interfaces, that does *not* define how those > > things > > > are to be implemented. This makes it easier to understand the > standard > > > and have separate (portable) libraries for the subsets, it does not > > > influence optimization possibilities by implementing everything in a > > > monolithic block (i.e., central progress). > > > > > > > - Both blocking & nonblocking belong to the core > > > > - Torsten > > > > - Some collectives may go into selectable subsets > > > I see three subsets: blocking colls, non-blocking colls and > > topological > > > colls (maybe also blocking / non-blocking). > > > > > > > - MPI_ANY_SOURCE considered harmful > > > I'd like to add datatypes and heterogeneity to this list (with > regards > > > to performance). Alexander mentioned the dynamics. I think we should > > > have a lit of items ready that could influence optimization > > > possibilities significanty if they were to be announced by the user > > > before he can use them. That would give another strong argument for > > the > > > subsetting. > > > > > > Best, > > > Torsten > > > > > > -- > > > bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ > ----- > > > Indiana University | http://www.indiana.edu > > > Open Systems Lab | http://osl.iu.edu/ > > > 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA > > > Lindley Hall Room 135 | +01 (812) 855-3608 > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > --------------------------------------------------------------------- > > > Intel GmbH > > > Dornacher Strasse 1 > > > 85622 Feldkirchen/Muenchen Germany > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > VAT Registration No.: DE129385895 > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > This e-mail and any attachments may contain confidential material > for > > > the sole use of the intended recipient(s). Any review or > distribution > > > by others is strictly prohibited. If you are not the intended > > > recipient, please contact the sender and delete all copies. > > > > > > > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > _______________________________________________ > > Mpi3-subsetting mailing list > > Mpi3-subsetting_at_[hidden] > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > --------------------------------------------------------------------- > > Intel GmbH > > Dornacher Strasse 1 > > 85622 Feldkirchen/Muenchen Germany > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > VAT Registration No.: DE129385895 > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > This e-mail and any attachments may contain confidential material for > > the sole use of the intended recipient(s). Any review or distribution > > by others is strictly prohibited. If you are not the intended > > recipient, please contact the sender and delete all copies. > > > > > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > > --------------------------------------------------------------------- Intel GmbH Dornacher Strasse 1 85622 Feldkirchen/Muenchen Germany Sitz der Gesellschaft: Feldkirchen bei Muenchen Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer Registergericht: Muenchen HRB 47456 Ust.-IdNr. VAT Registration No.: DE129385895 Citibank Frankfurt (BLZ 502 109 00) 600119052 This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.

1 0

Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon ww09
by unknown＠example.com 29 Feb '08

29 Feb '08

Alexander: Re: > Thanks. I understand your motivation. When you say "most real > applications" - what applications do you mean? At least, in what area? ? Scientific computing... > For the NIC part, the stress was on "here". In my opinion, subsetting is > not about making things more complicated, more challenging to the > implementors, or to the underlying hardware. It's about making things > simple, easy to use, and easy to implement - including implementation of > only those features your users actually need. That the implementation > may be faster due to this is an added bonus, not the primary goal. The emphasis here should not be on creating a disincentive for vendors to do the right thing... > Still, regarding user side copying. Yes, when people do this one wonders > why. There's a reason, apart from them: 1) not caring about datatypes > and their complexity and 2) not trusting their performance. A modern > compiler can rather well optimize a loop with a constant stride, and may > have difficulty with an unknown stride. This is why explicit loops are > sometimes indeed faster (much faster) in the resulting code than any > generic implementation. Huh? What makes you think the user copying code is in terms of constant stride? Generally, it varies with the input. We are not talking about a simple situation to optimize at the user level... Bronis > > Best regards. > > Alexander > > -----Original Message----- > From: Bronis R. de Supinski [mailto:bronis_at_[hidden]] > Sent: Friday, February 29, 2008 6:20 AM > To: Supalov, Alexander > Cc: mpi3-subsetting_at_[hidden] > Subject: RE: [Mpi3-subsetting] agenda for subsetting kickoff telecon > ww09 > > > Alexander: > > Most real applications need to send non-contiguous > data. If they do not use datatypes then they are > doing the equivalent of either the packing/unpacking > or smaller messages at the user level. This s hould > be discouraged, not encouraged. A small savings > in library object size is not ample reason to go > against that. And, yes, we are after encouraging > hardware vendors to provide the right hardware. > > Bronis > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > Hi, > > > > Thanks. I think the main thrust here is the library footprint (no > > pack/unpack, etc.) and complexity of the user side of the datatype > > interface, rather than performance. Many applications just don't need > > any of this, and never will. Why not translating this application > > non-requirement into a minimum MPI subset? Same with > communicator/group > > management, etc. > > > > Moreover, homogeneous installations that dominate HPC now don't > actually > > need any datatype support at all. They send chunks of bytes. This may > > change in the future, though. > > > > A minor performance implication is that without holes that are only > > possible with derived datatypes, one does not need to track this, > split > > the critical path, and make special provisions inside the MPI device > > layer to handle iov or such. > > > > The NIC capability argument is interesting, but it turns the > discussion > > on its head: we're not after motivating network vendors to provide > > scatter/gather in hardware here, are we? Please clarify. > > > > Best regards. > > > > Alexander > > > > -----Original Message----- > > From: mpi3-subsetting-bounces_at_[hidden] > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > Bronis > > R. de Supinski > > Sent: Friday, February 29, 2008 5:53 AM > > To: mpi3-subsetting_at_[hidden] > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > ww09 > > > > > > All: > > > > OK, I have to respond to the notion that derived datatypes > > limit performance. It is just not a reasonable position. > > > > Sure, if you can send contiguous locations, you will get > > higher performance. The problem is that codes do not only > > need to send contiguous data so that is not an adequate > > reason to say derived datatypes limit performance. > > > > So, what is left? That there is some more efficient way > > to send non-contiguous data? How? As multiple messages, > > each of which send contiguous data? If so, then the > > implementation could do this under the covers and the > > datatypes are just a convenience for the user not to > > have to specify the individual sends. OK, suppose that's > > not the reason. Perhaps the user can do the copying into > > a contiguous buffer and get better performance? While > > I have seen this hold with some implementations, it is > > absurd. There is no reason that I can discern as to why > > the user should be able to deduce a better copying > > mechanism than the MPI implementer. So, again, at worst, > > the datatypes should be a convenience. Do you have an > > alternative reason or a refutation of these opinions? > > > > What is more important, it is certainly possible to build > > scatter/gather support into a NIC and achieve better > > performance with datatypes than without. While there are > > issues to be resolved for that (primarily the issue of > > pinning memory), they are solvable with the right hardware > > mechanism. Just because it does not yet exist is not > > an adequate reason to say "Get rid of datatypes". OK, > > you are not saying that but you are saying to deprecate > > them in a sense. And saying you could send contiguous > > sends more efficiently is a bad argument here. How do > > datatypes cause inefficiency for that? How much is > > that cost really? At what point do you hit where the > > answer is "It would be faster not to compute anything"? > > > > Bronis > > > > > > On Fri, 29 Feb 2008, Supalov, Alexander wrote: > > > > > Hi, > > > > > > Thanks. What subsets inside the current standard would you propose? > > What > > > interfaces between them would you envision? > > > > > > Good idea about the optimization opportunities. Here's an initial > > > combined list, with the main benefits as I see them. Please > > > comment/extend. > > > > > > - Dynamic process support: less overhead in the progress engine, > > easier > > > global rank handling. > > > - Heterogeneity: better memory footprint, easier data handling. > > > - Derived datatypes (especially those with holes): better memory > > > footprint. > > > - MPI_ANY_SOURCE: faster, more simple multifabric progress. > > > - File I/O: smaller requests, easier wait/test functions. > > > - One-sided ops: no passive target w/o MPI calls - no extra progress > > > thread. > > > - Communicator & group management: better memory footprint. > > > - Message tagging: better support for stable dataflow exchanges, > > smaller > > > packets. > > > - Non-blocking communication: easier ordering, simplified request > > > handling. > > > > > > Best regards. > > > > > > Alexander > > > > > > -----Original Message----- > > > From: mpi3-subsetting-bounces_at_[hidden] > > > [mailto:mpi3-subsetting-bounces_at_[hidden]] On Behalf Of > > > Torsten Hoefler > > > Sent: Friday, February 29, 2008 5:08 AM > > > To: mpi3-subsetting_at_[hidden] > > > Subject: Re: [Mpi3-subsetting] agenda for subsetting kickoff telecon > > > ww09 > > > > > > Hi, > > > > Present: Leonid Meyerguz (Microsoft), Rich Graham (ORNL), > Richard > > > Barrett > > > > (ORNL), Torsten Hoefler (ISU), Alexander Supalov (Intel) > > > just for the record, it's "IU" not "ISU" :-) > > > > > > > - Scope of the effort > > > > - Rich > > > > - Minimum subset consistent with the rest of MPI, for > > > > performance/memory footprint optimization > > > > - Danger of splitting MPI, hence against optional features > in > > > the > > > > standard > > > I back that (danger of optional features for portability). I'd > propose > > > to split the current standard into mostly self-contained subsets > that > > > have clearly defined interfaces to the rest of the standard. Note: > > this > > > only defines logical interfaces, that does *not* define how those > > things > > > are to be implemented. This makes it easier to understand the > standard > > > and have separate (portable) libraries for the subsets, it does not > > > influence optimization possibilities by implementing everything in a > > > monolithic block (i.e., central progress). > > > > > > > - Both blocking & nonblocking belong to the core > > > > - Torsten > > > > - Some collectives may go into selectable subsets > > > I see three subsets: blocking colls, non-blocking colls and > > topological > > > colls (maybe also blocking / non-blocking). > > > > > > > - MPI_ANY_SOURCE considered harmful > > > I'd like to add datatypes and heterogeneity to this list (with > regards > > > to performance). Alexander mentioned the dynamics. I think we should > > > have a lit of items ready that could influence optimization > > > possibilities significanty if they were to be announced by the user > > > before he can use them. That would give another strong argument for > > the > > > subsetting. > > > > > > Best, > > > Torsten > > > > > > -- > > > bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ > ----- > > > Indiana University | http://www.indiana.edu > > > Open Systems Lab | http://osl.iu.edu/ > > > 150 S. Woodlawn Ave. | Bloomington, IN, 474045-7104 | USA > > > Lindley Hall Room 135 | +01 (812) 855-3608 > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > --------------------------------------------------------------------- > > > Intel GmbH > > > Dornacher Strasse 1 > > > 85622 Feldkirchen/Muenchen Germany > > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > > VAT Registration No.: DE129385895 > > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > > > This e-mail and any attachments may contain confidential material > for > > > the sole use of the intended recipient(s). Any review or > distribution > > > by others is strictly prohibited. If you are not the intended > > > recipient, please contact the sender and delete all copies. > > > > > > > > > _______________________________________________ > > > Mpi3-subsetting mailing list > > > Mpi3-subsetting_at_[hidden] > > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > > > > _______________________________________________ > > Mpi3-subsetting mailing list > > Mpi3-subsetting_at_[hidden] > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-subsetting > > --------------------------------------------------------------------- > > Intel GmbH > > Dornacher Strasse 1 > > 85622 Feldkirchen/Muenchen Germany > > Sitz der Gesellschaft: Feldkirchen bei Muenchen > > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > > VAT Registration No.: DE129385895 > > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > > > This e-mail and any attachments may contain confidential material for > > the sole use of the intended recipient(s). Any review or distribution > > by others is strictly prohibited. If you are not the intended > > recipient, please contact the sender and delete all copies. > > > > > --------------------------------------------------------------------- > Intel GmbH > Dornacher Strasse 1 > 85622 Feldkirchen/Muenchen Germany > Sitz der Gesellschaft: Feldkirchen bei Muenchen > Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer > Registergericht: Muenchen HRB 47456 Ust.-IdNr. > VAT Registration No.: DE129385895 > Citibank Frankfurt (BLZ 502 109 00) 600119052 > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > >

1 0