January 2012 - Asg - mailman3.cels.anl.gov

Asg page Admin test page has been created by Carns
by WikiAdmin 26 Jan '12

26 Jan '12

Dear Asg-mailing-list, The Asg page Admin test page has been created on 21:23, 26 January 2012 by Carns, see https://wiki.mcs.anl.gov/asg/index.php/Admin_test_page for the current revision. This is a new page. Editor's summary: Testing mailing list notifications on wiki change Contact the editor: mail: https://wiki.mcs.anl.gov/asg/index.php/Special:EmailUser/Carns wiki: https://wiki.mcs.anl.gov/asg/index.php/User:Carns There will be no other notifications in case of further changes unless you visit this page. You could also reset the notification flags for all your watched pages on your watchlist. Your friendly Asg notification system -- To change your watchlist settings, visit https://wiki.mcs.anl.gov/asg/index.php/Special:Watchlist/edit Feedback and further assistance: https://wiki.mcs.anl.gov/asg/index.php/Help:Contents

1 0

Asg-private e-mail address confirmation
by carns＠mcs.anl.gov 26 Jan '12

26 Jan '12

Someone, probably you, from IP address 24.30.66.155, has registered an account "Asg-mailing-list" with this e-mail address on Asg-private. To confirm that this account really does belong to you and activate e-mail features on Asg-private, open this link in your browser: https://wiki.mcs.anl.gov/asg-private/index.php/Special:ConfirmEmail/38eca5f… If you did *not* register the account, follow this link to cancel the e-mail address confirmation: https://wiki.mcs.anl.gov/asg-private/index.php/Special:Invalidateemail/38ec… This confirmation code will expire at 21:17, 2 February 2012.

1 0

security text for compendium
by Dries Kimpe 25 Jan '12

25 Jan '12

I've added the security text for the compendium... https://wiki.mcs.anl.gov/asg-private/index.php/Kickoff-2012-day2#Security Dries

1 0

SSM no-unlink flag, pseudocode, notes
by Matthew Farmer 24 Jan '12

24 Jan '12

All, As requested, you can expect a flag to support the unlink semantics discussed -- SSM_ME_NEVER_UNLINK. (This is not the same as SSM_ME_AUTO_RELINK, which has the match entry in a blocked state until the transfer is done.) That is a demand to the API that the match buffer remain available for reading and writing explicitly until the buffer is unlinked. I'll get support for that ASAP after the level-1 next week. There is an untested existing flag SSM_ME_MATCH_STATIC which is intended to serve the same purpose, and it will likely merely be renamed to the new flag after testing. You are welcome to try it, but it's again not tested. *grin* Here's a rough sketch of SSM operation with the new flag; this code does a server-managed transfer of some arbitrarily-described buffer: SERVER startup: ssm_start(desired_transport, ...) // link a buffer for requesting data datareq_descriptor_buffer_me = ssm_link(datareq_descriptor_buffer, datareq_callback, datareq_match_bits, SSM_ME_NEVER_UNLINK, ...) loop ssm_wait datareq_callback: // load requested data, post, and notify databuf = load_data(datareq_descriptor_buffer) databuf_get_info = setup_get(databuf); databuf_me = ssm_link(databuf, databuf_callback, databuf_get_info->match_bits, SSM_ME_NEVER_UNLINK, ...) databuf_get_info_me = ssm_link(databuf_get_info, databuf_get_info_cb, get_info_match_bits, SSM_ME_NEVER_UNLINK, ...) ssm_put(datareq_descriptor_buffer->client, databuf_get_info_me) databuf_get_info_cb: ssm_unlink(databuf_get_info_me) databuf_callback: // client got the data ssm_unlink(databuf_me) shutdown: ssm_unlink(datareq_descriptor_buffer_me) ssm_stop CLIENT get_data: descriptor = data_load_descriptor(...) descriptor_me = ssm_link(descriptor, descriptor_callback, datareq_match_bits, SSM_ME_NEVER_UNLINK, ...) databuf_get_info_me = ssm_link(databuf_get_info, databuf_get_info_cb, get_info_match_bits, SSM_ME_NEVER_UNLINK, ...) ssm_put(server, descriptor_me) descriptor_callback: // put request to server ssm_unlink(descriptor_me) databuf_get_info_cb: // server told us how to get requested data ssm_unlink(databuf_get_info_me) ssm_link(databuf, databuf_cb, databuf_get_info->match_bits, SSM_ME_NEVER_UNLINK, ...) ssm_get(server, databuf); databuf_cb: // got the data ssm_unlink(databuf) END EXAMPLE Some notes: - The callbacks in this example are simplified; when running callbacks will also get INIT events when operations start, and you of course wouldn't want to unlink as a response to those. Any other event type is a completion and can be treated like one. - SSM addresses generally live on the heap, and the semantics for generated ones are screwy right now, which is to say wrong with serious ownership issues. Right now if the client doesn't free the provided address for any event that's not a local init it will leak. Fixing has gotten rolled in with the new match list semantics, and so is still exposed. Suggested fix is linking entries with the SSM_ME_AUTO_FREE_ADDR flag as long as the addr doesn't need to persist; this flag will free the provided addr after the callback completes, if needed. That flag (and the other FREE flags) will disappear when I fix the memory semantics. (In short, currently you need to free the provided remote_addr for events that are not both SSM_INIT and OWNER_LOCAL, which that flag will do automatically.) - All the flags in ssmdef.h operate as described, if they sound handy. Some of them are a bandaid for busted memory semantics and will disappear with the new match list stuff. Also, as discussed, match list and buffer semantics are shifting around, and a side effect of that will be supporting the operation modes that were described at the meeting. In existing implementations these changes will replace and/or augment ssm_link and ssm_unlink commands and shouldn't affect much of the rest of the code. I'll write up a transition guide as well as the new semantics when I finish them *grin*. Does everyone have the latest SSM drop? The SSM portion of the Nebula repo is getting reorganized a bit and pulling together that drop might be a bit of a pain whilst I'm shifting stuff around, so it may be worth keeping handy until the next drop is ready. (The tarball of that drop is not moving -- just the svn:externals links are going away, so that it may take more than one checkout to build an old drop.) Flag will be live soon after next week, with new match list stuff soon after. Hope that helps and please let me know if you have any questions or concerns. Shane

2 3

wiki accounts updated
by Phil Carns 24 Jan '12

24 Jan '12

FYI, I've added everyone in the lists provided by Haiying, Richard, and Shane to the wikis as of a few minutes ago. -Phil

1 0

Current membership list
by Dries Kimpe 23 Jan '12

23 Jan '12

As discussed on the call, the current subscriber list is below. (sorted by organisation to simplify validation). Let me know if people need to be added or removed. Thanks, Dries carns(a)mcs.anl.gov dkimpe(a)mcs.anl.gov harms(a)alcf.anl.gov rross(a)mcs.anl.gov rbbrigh(a)sandia.gov rklundt(a)sandia.gov lee(a)sandia.gov mlcurry(a)sandia.gov tony(a)cis.uab.edu efrees(a)cis.uab.edu mgperez(a)uab.edu zhwsun(a)uab.edu rrb(a)acm.org settlemyerbw(a)ornl.gov gemmill(a)clemson.edu shenh(a)clemson.edu syun(a)g.clemson.edu pooja.bharath.shah(a)gmail.com shanefarmer22(a)gmail.com

4 3

Call for Workshops at IEEE eScience, due January 23, 2012
by Jill Gemmill 23 Jan '12

23 Jan '12

This is due today, but --- is anyone on this list already involved in organizing a session on exascale file systems? If not, should we propose something? Discuss on phone call? By the way, when will the call-in number be posted? n Jill Call for Workshops 8th IEEE International Conference on eScience October 8-12, 2012 Chicago, IL, USA The 8th IEEE eScience conference (e-Science 2012), sponsored by the IEEE Computer Society's Technical Committee for Scalable Computing (TCSC), will be held in Chicago Illinois from 8-12th October 2012. The eScience 2011 conference is designed to bring together leading international and interdisciplinary research communities, developers, and users of eScience applications and enabling IT technologies. Multiple e-Science 2012 Workshops will be held on Monday and Tuesday, 8th and 9th October, co-located with the main conference. Workshops are an important part of the conference in providing opportunity for researchers to present their work in a more focused way than the conference itself and to have discussion of particular topics of interest to the community. We cordially invite you to submit workshop proposals on any eScience related topic to the Workshop Chair. To help those interested know their purpose and scope, workshop proposals should include: * A description of the workshop, its focus, goals, and outcome * A draft call for papers * Names and affiliations of the organizers and tentative composition of the committees * Expected numbers of submissions and accepted papers * Prior history of this workshop, if any. Please include: number of submissions, number of accepted papers, and attendee count. Workshop organizers are responsible for establishing a program committee, collecting and evaluating submissions, notifying authors of acceptance or rejection in due time, ensuring a transparent and fair selection process, organizing selected papers into sessions, and assigning session chairs. Proposals will be selected that show clear focus and objectives in areas of emerging or developing interest guaranteed to generate significant interest in the community. Once accepted, the workshop should establish its own paper submission system. For each paper selected for publication, an author must be registered for eScience 2012. Each paper must be presented in person by at least one of the authors. It is expected that the proceedings of the eScience 2012 workshops will be published by the IEEE Computer Society Press, USA and will be made available online through the IEEE Digital Library. SUBMISSION PROCESS Workshop proposals should be emailed to escience2012-workshops(a)fnal.gov<mailto:[email protected]?subject=Workshop%20Submission%20for%20the%208th%20IEEE%20International%20Conference%20on%20eScience> ORGANIZATION General Chair * Ian Foster, University of Chicago & Argonne National Laboratory, USA Program Co-Chairs * Daniel S. Katz, University of Chicago & Argonne National Laboratory, USA * Heinz Stockinger, SIB Swiss Institute of Bioinformatics, Switzerland Workshops Chair * Ruth Pordes, FNAL, USA Sponsorship Chair * Charlie Catlett, Argonne National Laboratory, USA Conference Manager and Finance Chair * Julie Wulf-Knoerzer, University of Chicago & Argonne National Laboratory, USA Publicity Chairs * Kento Aida, National Institute of Informatics, Japan * Ioan Raicu, Illinois Institute of Technology, USA * David Wallom, Oxford e-Research Centre, UK Local Organizing Committee * Ninfa Mayorga, University of Chicago, USA * Evelyn Rayburn, University of Chicago, USA * Lynn Valentini, Argonne National Laboratory, USA -- ================================================================= Ioan Raicu, Ph.D. Assistant Professor, Illinois Institute of Technology (IIT) Guest Research Faculty, Argonne National Laboratory (ANL) ================================================================= Data-Intensive Distributed Systems Laboratory, CS/IIT Distributed Systems Laboratory, MCS/ANL ================================================================= Cel: 1-847-722-0876 Office: 1-312-567-5704 Email: iraicu(a)cs.iit.edu<mailto:[email protected]> Web: http://www.cs.iit.edu/~iraicu/ Web: http://datasys.cs.iit.edu/ ================================================================= =================================================================

2 1

Aesop project access
by Phil Carns 23 Jan '12

23 Jan '12

We've set up public keys from Matthew, Tony, Amin, Zhiwei, Jeff, and Seok that will allow you to commit to the aesop repository. Let me know if I missed anyone, check the wiki for instructions on how to do an authenticated checkout, and of course let us know if you have any problems: https://trac.mcs.anl.gov/projects/aesop It ended up being harder than we thought to add accounts for the trac site itself (long story). Instead we've decided to set things up this way: - git repo: anonymous read access for everyone, read/write access for folks that give us their public key - trac tickets: allows anonymous submissions, and you can add your email address if you want to get email notification on specific ticket changes - trac wiki: please send your edits/suggestions to Dries, Phil, or Kevin I also wanted to point out that you can subscribe to an rss feed for aesop here if you want to see any wiki, ticket, or code revision changes: https://trac.mcs.anl.gov/projects/aesop/timeline thanks, -Phil

1 0

Conference call information
by Ward, Lee 22 Jan '12

22 Jan '12

Is available on the private wiki page. Next call is Monday, the 23rd, at 2 PM Mountain time. --Lee

1 0

project wikis and draft compendium
by Phil Carns 22 Jan '12

22 Jan '12

Hi everyone, We have two wikis ready at the following urls: https://wiki.mcs.anl.gov/asg (readable by anyone) https://wiki.mcs.anl.gov/asg-private (readable only if you log in) I'll add accounts for everyone shortly, so keep an eye out for two emails asking you to log in and reset your password. Let me know if you don't see them by Monday morning. The asg-private wiki has a starting point for the kickoff meeting compendium. I uploaded all of the presentations in pdf format and added some discussion notes from all of the sessions that I witnessed. Maybe this is too verbose? Please correct, edit, or reformat as needed. The architecture breakout session, security breakout session, and revised milestone section are all empty right right now. I don't have any notes for that stuff. I think we want to have the compendium done by Friday. thanks, -Phil

1 0