- ogsa-bes-wg - lists.ogf.org

Process issues
by Stephen M Pickles 14 Feb '07

14 Feb '07

All, One week last call on the OGSA-BES spec was announced about 10 days ago. This has sparked a flurry of interesting discussion. However, I think it time that the group try to reign in these discussions, and decide what if anything needs to be done to the document before submitting it to the OGF editor. I see some cause for urgency, as the OGSA-HPCP Profile specification, which references OGSA-BES, is already in GFSG review and on-track to emerge for public comment as a proposed recommendation in two weeks' time. The sub-threads I have noticed are: 1. A discussion about authorship and contributors, which should not be hard for grown-ups to reach closure on. Remember that the limit of 5 on authors is purely advisory. I recommend giving credit where credit is due, and if you still have more than 5 authors, submit anyway, making it the GFSG's problem, not the group's. 2. A discussion about the right granularity for BES faults. I don't detect any profound controversy here (just a bunch of smart people trying to get it right), but on the other hand, I don't detect a clear consensus on how to proceed. 3. Some arcane discussion, apparently stillborn, on Weak Similarity and the state model, which I don't presume to understand 4. A set of questions from Joseph Bester, of which some appear to have been answered, and some don't. May I suggest that the group has a call to try to close off these loose ends? And in the meantime, could anyone who still has issues with the current draft propose some concrete text? Stephen ==================== Stephen M. Pickles ==================== Technical Director, National Grid Service Manchester Computing Room G49.1, Kilburn Building The University of Manchester tel: +44 161 275 5974 Oxford Road fax: +44 161 275 6800 Manchester M13 9PL stephen.pickles(a)manchester.ac.uk

1 0

BES Last Call
by Mark Morgan 14 Feb '07

14 Feb '07

Ladies and Gentlemen, As per consensus at the BES session yesterday, 1 February 2007, at 6pm at OGF19, I have made all noted modifications to the BES document. It is uploaded to the Gridforge website and is available as ogsa-bes-draft-v30. If you feel that you deserve acknowledgements for work done, or to be listed as an author, and you are not currently indicated properly in the document, please speakup. Once again, this is the last call for OGSA BES specification before we send it off for public comment. -Mark -- Mark Morgan Research Scientist Department of Computer Science University of Virginia http://www.cs.virginia.edu mmm2a(a)virginia.edu (434) 982-2047

8 18

New document
by Mark Morgan 12 Feb '07

12 Feb '07

Apparently I greatly offended Mr. Ian Foster who felt that his position on the author list didn't represent either his effort, or the IETF rules (which until today I didn't know). I have uploaded a new document which alphabatizes the names and have removed myself from the author list. Please see document version 31, now located in gridforge. -Mark

6 6

Last call
by Andrew Grimshaw 29 Jan '07

29 Jan '07

All, Last call on the version uploaded by Chris on 1/24 starts today. We will discuss issues at the session this week. The ideal is to get it to the AD's in a week, so it can go to the GFSG and then to public comment. A Andrew Grimshaw Professor of Computer Science University of Virginia 434-982-2204 grimshaw(a)cs.virginia.edu

1 0

invitation to OGSA information model session, Tues Jan 30 at 11am
by Ellen Stokes 26 Jan '07

26 Jan '07

BES and HPC folks, Please plan to come and participate in the OGSA Information & Data Modeling session on Tues Jan 30, 11am-12:30pm, in the Bellflower room. As you recall, a decision of the BES workgroup was to remove the information model section from the BES specification. Since then, a separate draft has been written to contain that material and utlimately publish it as an OGF Informational document that can be forwarded to DMTF for inclusion in the Common Information Model. The OGSA resource management design team will be reviewing this document at this session - and we'd like your input. ( https://forge.gridforum.org/sf/go/doc14004?nav=1) We'll also be discussing a proposal for information/data modeling using XML and XQuery. See OGF19 schedule for complete agenda for this session. Ellen Ellen Stokes STSM, Grid Computing Member, IBM Academy of Technology Austin, Texas tl 678 0552 outside +1 512 838 0552 stokese(a)us.ibm.com

1 0

BES specification update
by Christopher Smith 24 Jan '07

24 Jan '07

Hi all, I've uploaded a new version of the BES specification to Gridforge. I've put a change summary in the version notes. -- Chris

1 0

regarding BESActivity wsdl
by Ahmed Shiraz Memon 09 Jan '07

09 Jan '07

Hi, May i ask for the correct version of WSDL of BES Activty, since the latest version i.e. 28 has wsdl but having some errors and missing tags which might be bit of confusion. Any help would be great. Thanks alot in advance, Shiraz

1 0

Latest draft
by Andrew Grimshaw 02 Jan '07

02 Jan '07

All, I have uploaded to grid forge version 28, an update of v26 with the modifications called for in the September 13 F2F minutes. There are still a few changes required. My recollection is that Chris Smith now has the pen to fix the WSDL and examples to match the HPC profile results. Can we schedule a telecom next week? A Andrew Grimshaw Professor of Computer Science University of Virginia 434-982-2204 grimshaw(a)cs.virginia.edu

2 1

GetActivityDocuments doesn't
by Peter G. Lane 20 Dec '06

20 Dec '06

In trying to finish up some stuff before I leave I noticed a discrepancy in the factory schema. The GetActivityDocumentsResponseType doesn't in fact return ActivityDocumentTypes. Here is the message type: <xsd:complexType name="GetActivityDocumentsResponseType"> <xsd:sequence> <xsd:element name="Response" type="bes-factory:GetActivityDocumentResponseType" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> And here is the response array element type: <xsd:complexType name="GetActivityDocumentResponseType"> <xsd:sequence> <xsd:element ref="bes-factory:ActivityIdentifier"/> <xsd:element ref="jsdl:JobDefinition" minOccurs="0"/> <xsd:element name="Fault" type="xsd:anyType" minOccurs="0"/> </xsd:sequence> </xsd:complexType> The activity documents are more than just the JobDefinition: <xsd:complexType name="ActivityDocumentType"> <xsd:sequence> <xsd:element ref="jsdl:JobDefinition"/> <xsd:any namespace="##other" processContents="lax" minOccurs="0" maxOccurs="unbounded"/> </xsd:sequence> </xsd:complexType> The ref to jsdl:JobDefinition ought to be a ref to ActivityDocument as so: <xsd:complexType name="GetActivityDocumentResponseType"> <xsd:sequence> <xsd:element ref="bes-factory:ActivityIdentifier"/> <xsd:element ref="bes-factory:ActivityDocument" minOccurs="0"/> <xsd:element name="Fault" type="xsd:anyType" minOccurs="0"/> </xsd:sequence> </xsd:complexType> Just returning the JobDefinition means that any 'any's, such as the subscription or idempotent creation ID extensions, in the ActivityDocuments used in CreateActivity will be irretrievable. Peter

1 0

Good Bye
by Peter G. Lane 15 Dec '06

15 Dec '06

Hi All, Some recent circumstances have forced me to quit my job and become a stay-at-home dad for a year. I will probably monitor things from time to time, but at this point I don't have any plans to continue my involvement in the BES and HPCP working groups. Participating in these working groups has been a lot of fun as well as a useful experience. I learned a lot. It was nice meeting those of you who I did meet at OGF, and I regret missing many of the other HPCP demo participants at SC. Good luck to everyone and the future of open standards. Peter P.S. Please contact Stu Martin (smartin(a)mcs.anl.gov) if you need to ask any questions about the Globus Project's BES implementation/plans.

1 0