====================================================================== ACS Teleconference ----------------------------------------------------------------------- Date and Time: March 30 19:00-20:30 EST 2005 / March 31 9:00-10:30 JST, 2005 Participants: Keisuke Fukui (Fujitsu) Peter Ziu (Northrop) Michael Behrens (R2AD) Sachiko Wada (ASCADE) - minutes Apologies: Thomas Studwell (IBM) ------------------------------------------------------------------------ * Topics a) the new requirements from NAREGI PSE + Keisuke summarized some new requirements for ACS spec are mentioned in the GGF13 presentation at ACS #1 session, which include: - including source code to be compiled for each target platform at or prior to the deployment of the application. - storing compiled binaries for target platforms in the repository. - storing (caching) the input and output data pairs through the lifecycle of the multiple executions of the application. + Agreed that we need to discuss about the above more in detail. Below are some additional issues raised in the discussion. - If source code is dynamically compiled at runtime, who is responsible for assuring the compiler in place? - Since the data used in scientific calculation tend to be significantly large, repositories for dynamic data may be better if implemented separately from the ACS repository. + Actions - Study the presentation and point out more questions/observations. b) relationship to the deployment API and component model being discussed in the CDDLM-WG. + Keisuke reported that the CDDLM team is positive about seeking for collaboration with ACS. They were preparing three specifications to be completed and one of them was the deployment API, whose 'addFile' interface looks to be a candidate contact point with ACS, for example, "register" interface of ARI. We need to discuss about where the ACS interfaces should stand in the CDDLM's lifecycle model. + Agreed that we also seek for the collaboration with the CDDLM-WG. The ACS repository will make it easy to enable versioning and repeatable deployment. We need to learn their specs. + Actions - Study their specification and point out some questions/observations. c) collaboration with IUDD standardization group + Generic archive standard will be output from IUDD standardization and we agreed that grid specific archive format (AAF) will use IUDD as a framework. We need to discuss differences in requirements between IUDD(for generic use) and AAF(grid specific) and the way to utilize IUDD spec so that it is efficient for grid use. + Actions - Make clear our (grid) requirements for IUDD and discuss with IUDD team. d) recruiting involvement from the European communities. + Keisuke reminded us that one of the comments I got after the sessions is that we should seek for the participants from European Grid communities since they share a lot with our usecase and scope. We need a research on this possibility. UniGrid, EGEE, omii, and more... + We need researches on their activity to examine if there exists any possibility of collaboration or integration with our activity. It is expected to promote our activity. However, since we should first establish collaboration with NAREGI, CDDLM and SI. + Actions - Keep watching other activities which might have a potential relationship to ACS * Roadmap a) Working draft We need to move forward toward writing a work group draft, which is planned to be presented at GGF14 in the charter . b) implementation The Business Grid and NAREGI projects are expected to be implementation of ACS spec. To realize this, we should write up our specification quickly. On the other hand, since their products are rather large scale, so we may seek for some more concise implementations to refine our concept. * F2F meeting plan Not decided, but one possible candidate is: Date: 2nd week of May Place: Washington, DC Waiting for the response from Tom and other participants. * Next call Date(Pending): Apr 20 Wed 20:00 EST/Apr 21 Thu 10:00 JST # U.S. Daylight Saving time has started Apr 3.