Re: [Pgi-wg] Sec: Agreement on WS-Naming (ref by strawman)
Ok, I understand much better know. So I see a demand that we discuss if our services are contacted with WS-Addressing or not. Q: How does ARC services then distinguish one instance from another if only one URL is used?! you don't seem to use referenceproperties then, instead adding to the url resource-instance ids?! Take care, Morris -------------------------------------------------------------------------------- Morris Riedel SW - Engineer Distributed Systems and Grid Computing Division Central Institute of Applied Mathematics Research Centre Juelich Wilhelm-Johnen-Str. 1 D - 52425 Juelich Germany Email: m.riedel@fz-juelich.de Info: http://www.fz-juelich.de/zam/ZAMPeople/riedel Phone: +49 2461 61 - 3651 Fax: +49 2461 61 - 6656 Skype: MorrisRiedel 'We work to improve ourselves and the rest of mankind.' ----- Original Message ----- From: Aleksandr Konstantinov <aleksandr.konstantinov@fys.uio.no> Date: Friday, March 20, 2009 12:44 pm Subject: Re: [Pgi-wg] Sec: Agreement on WS-Naming (ref by strawman)
On Friday 20 March 2009 12:24, Morris Riedel wrote:
Hi,
- As I -explained in an earlier mail, "limited" here means that the PGI profile -shall precisely described in a self-contained way the minimum set of EPR -elements which must be supported by conforming inplementations.
So far, I guess we only depend on the wsa:To element of the specification> with the URI making it the minimum element of it.
With WS-Naming this would be expanded using wsa:metadata and ws- naming> elements of course.
Isn't our usual procedure to take the URI - encode an EPR (with wsa:To) and send it to the BES systems? Or is this maybe only my narrow UNICORE view of how we do it?
Q: What is meant by accessing a CREAM-BES and A-REX with URI (instead of EPRs)?
ARC does not rely on EPR present in SOAP messages header sent to BES service. ARC clients do put EPRs into headers for compatibility. Hosting environment of ARC services may perform routing of messages depending on wsa:To element of EPR. To make it short - we can live with and without EPRs. AFAIR BES specs say nothing about if BES service has to be contacted using EPRs.
A.K. _______________________________________________ Pgi-wg mailing list Pgi-wg@ogf.org http://www.ogf.org/mailman/listinfo/pgi-wg
------------------------------------------------------------------- ------------------------------------------------------------------- Forschungszentrum Juelich GmbH 52425 Juelich Sitz der Gesellschaft: Juelich Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 Vorsitzende des Aufsichtsrats: MinDir'in Baerbel Brumme-Bothe Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. Vorsitzender), Prof. Dr. Harald Bolt, Dr. Sebastian M. Schmidt ------------------------------------------------------------------- -------------------------------------------------------------------
So I see a demand that we discuss if our services are contacted with WS-Addressing or not.
In case this thread turns into a referendum on the merits of WS-Addressing, I would like to point out that that the Basic Execution Service (BES) spec incorporates EPRs into its interface. So, if PGI has apsirations to discuss anything that resembles BES, support for WS-Addressing would be a Good Idea if for no other other reason that to not have to completely re-invent the BES wheel (again). -Duane
On Friday 20 March 2009 16:22, you wrote:
So I see a demand that we discuss if our services are contacted with WS-Addressing or not.
In case this thread turns into a referendum on the merits of WS-Addressing, I would like to point out that that the Basic Execution Service (BES) spec incorporates EPRs into its interface.
But not for communication. EPRs in BES are for identification of activities. That could be any opaque string as well. Or any opaque string wrapped in EPR :) A.K.
So, if PGI has apsirations to discuss anything that resembles BES, support for WS-Addressing would be a Good Idea if for no other other reason that to not have to completely re-invent the BES wheel (again).
-Duane
Sorry for delay. On Friday 20 March 2009 15:57, m.riedel@fz-juelich.de wrote:
Ok,
I understand much better know.
So I see a demand that we discuss if our services are contacted with WS-Addressing or not.
Q: How does ARC services then distinguish one instance from another if only one URL is used?! you don't seem to use referenceproperties then, instead adding to the url resource-instance ids?!
Normally in ARC one instance of service corresponds to one URL. ARC services are not migrating. In ARC framework if wsa:To is present in request it will be used for routing message to corresponding service. If not - no problem - URL provided by underlying protocol (like HTTP) will be used. ARC clients use URL on user/human side (visual representation). A.K.
Good morning Aleksandr, thanks for the input - that's rather clear - but how you distinguish one 'job instance' from another if not with WS-Addressing EPRs? Thanks, Morris ------------------------------------------------------------ Morris Riedel SW - Engineer Distributed Systems and Grid Computing Division Jülich Supercomputing Centre (JSC) Forschungszentrum Juelich Wilhelm-Johnen-Str. 1 D - 52425 Juelich Germany Email: m.riedel@fz-juelich.de Info: http://www.fz-juelich.de/jsc/JSCPeople/riedel Phone: +49 2461 61 - 3651 Fax: +49 2461 61 - 6656 Skype: MorrisRiedel "We work to better ourselves, and the rest of humanity" Sitz der Gesellschaft: Jülich Eingetragen im Handelsregister des Amtsgerichts Düren Nr. HR B 3498 Vorsitzende des Aufsichtsrats: MinDirig'in Bärbel Brumme-Bothe Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. Vorsitzender)
------Original Message----- -From: Aleksandr Konstantinov [mailto:aleksandr.konstantinov@fys.uio.no] -Sent: Friday, March 27, 2009 9:01 AM -To: m.riedel@fz-juelich.de -Cc: Aleksandr Konstantinov; pgi-wg@ogf.org -Subject: Re: [Pgi-wg] Sec: Agreement on WS-Naming (ref by strawman) - - Sorry for delay. - -On Friday 20 March 2009 15:57, m.riedel@fz-juelich.de wrote: -> Ok, -> -> I understand much better know. -> -> So I see a demand that we discuss if our services are contacted with WS- -Addressing or not. -> -> Q: How does ARC services then distinguish one instance from another if only one -URL is used?! you don't seem to -> use referenceproperties then, instead adding to the url resource-instance ids?! - -Normally in ARC one instance of service corresponds to one URL. ARC services are -not migrating. -In ARC framework if wsa:To is present in request it will be used for routing message -to corresponding -service. If not - no problem - URL provided by underlying protocol (like HTTP) will be -used. -ARC clients use URL on user/human side (visual representation). - - -A.K.
participants (4)
-
Aleksandr Konstantinov -
Duane Merrill -
m.riedel＠fz-juelich.de -
Morris Riedel