[ogsa-wg] [glue-wg] OGSA Information Services

Omer F. Rana o.f.rana at cs.cardiff.ac.uk
Fri Aug 17 08:00:18 CDT 2007


Hi Laurence,

Have you looked at the discovery mechanisms found in peer-2-peer systems 
-- which are much
more likely to scale? There has been work in integrating P2P concepts 
with directory services
also.

My understanding of SAGA effort is that they identify that a discovery 
service is necessary -- but not
how the actual discovery mechanism works (as this is outside, surely, 
the scope of SAGA)?

regards
Omer


Laurence Field wrote:
> Hi Gerson,
>
> A common information system client would bets fit will with the SAGA 
> work. They are already working on an API for Service discovery which 
> will use plugins to find information from the different grids.
>
> The OGSA Information Services group is currently looking it existing 
> technologies used in information systems. Your experiences with MDS4 and 
> XPath would be valuable input to this group.  Do you have anything 
> available which documents your experiences?
>
> Thanks
>
> Laurence
>
>
> Gerson Galang wrote:
>   
>> Hi all,
>>
>> I don't have anything against having another layer sitting on top of 
>> all different Grid information systems implementation but I'm just not 
>> sure if that model will scale.
>>
>> I got interested in this discussion because I'm developing an 
>> application which is somewhat related to what I'm going to recommend 
>> below. The application is being developed for the Australian Grid 
>> users and developers and is going to be used in querying an MDS4 index 
>> service which uses the GLUE 1.2 schema. We have specified our own 
>> query language for querying the information system because we didn't 
>> want developers to write XPath queries if they directly interact with 
>> MDS4. As you know a very simple resource condition in an info systems 
>> query can turn out to be a very long query string when converted into 
>> XPath especially XPath 1.1 which is what MDS4 only understands.
>>
>> This group should probably look into implementing a similar generic 
>> Grid info systems client application for all the Grid info systems 
>> implementations around the world. Since the plan is to use the GLUE 
>> schema, there won't be any issues with specifying a standard query 
>> language that users can use with this client. The client will need to 
>> be able to query a "registry of endpoints" service to get information 
>> about the endpoint, query implementation (eg. wsrf/mds4, mds2, etc), 
>> and the query language (xpath, xquery, ldap, sql) to use for the info 
>> systems. The client will be the one in charge of transforming the 
>> user's query into the respective query languages used by the different 
>> Grid information systems implementations. I think we'll get a quicker 
>> response if we query the "registry of endpoints" service than the top 
>> level information system which indexes all the Grid resource 
>> information around the world.
>>
>> Our application can only handle a very simple query at the moment. See 
>> link below..
>>
>> https://www.grid.apac.edu.au/repository/svn/infosystems/GlueSchemaQueryLib/README 
>> <https://www.grid.apac.edu.au/repository/svn/infosystems/GlueSchemaQueryLib/README>
>>
>> but we have plans of extending the language later on to support 
>> queries like "(Site.Latitude > 0 and Site.Latitude < 50)". Composing a 
>> query for the GLUE schema shouldn't be that hard. The only thing which 
>> makes the query complicated is the language we use in querying the 
>> Grid info systems.
>>
>> I'm just not sure if what we are implementing is something you would 
>> want to look into.
>>
>> Regards,
>> Gerson
>>
>> On 8/4/07, *Andreas Savva* <andreas.savva at jp.fujitsu.com 
>> <mailto:andreas.savva at jp.fujitsu.com>> wrote:
>>
>>     The presentation Ravi is referring to is
>>     https://forge.gridforum.org/sf/go/doc14456?nav=1
>>     <https://forge.gridforum.org/sf/go/doc14456?nav=1>
>>
>>     Subramaniam, Ravi wrote:
>>     > Hi Hiro/Felix,
>>     >
>>     > Currently the examples in the document are from "academia" (not
>>     that I
>>     > am against that as most who know me will attest) but in the
>>     spirit of
>>     > including enterprise examples would it make sense to include the
>>     > implementation that I presented at the OGSA F2F in Manchester as an
>>     > example for the enterprise case?
>>     >
>>     > Ravi
>>     >
>>     > -----Original Message-----
>>     > From: ogsa-wg-bounces at ogf.org <mailto:ogsa-wg-bounces at ogf.org>
>>     [mailto:ogsa-wg-bounces at ogf.org <mailto:ogsa-wg-bounces at ogf.org>]
>>     On Behalf
>>     > Of Hiro Kishimoto
>>     > Sent: Wednesday, August 01, 2007 2:49 PM
>>     > To: ogsa-wg
>>     > Cc: glue-wg at ogf.org <mailto:glue-wg at ogf.org>
>>     > Subject: Re: [ogsa-wg] OGSA Information Services
>>     >
>>     > Hi all,
>>     >
>>     > Per Felix's request, I've upload revised draft onto the GridForge.
>>     >
>>     > https://forge.gridforum.org/sf/go/doc14692
>>     >
>>     > We will discuss it at Thursday's call.
>>     >
>>     > Thanks,
>>     > ----
>>     > Hiro Kishimoto
>>     >
>>     > -------- Original Message  --------
>>     > Subject: Re:[ogsa-wg] OGSA Information Services
>>     > From: Hiro Kishimoto < hiro.kishimoto at jp.fujitsu.com
>>     <mailto:hiro.kishimoto at jp.fujitsu.com> >
>>     > To: ogsa-wg <ogsa-wg at ogf.org <mailto:ogsa-wg at ogf.org>>
>>     > Cc: felix.ehm at cern.ch <mailto:felix.ehm at cern.ch>
>>     > Date: 2007/07/26 12:55
>>     >
>>     >> Hi all,
>>     >>
>>     >> Discussion material for Information Services by Laurence Field and
>>     >> Felix Ehm is now on the GridForge.
>>     >>
>>     >> https://forge.gridforum.org/sf/go/doc14692
>>     <https://forge.gridforum.org/sf/go/doc14692>
>>     >>
>>     >>  > This document covers the problem description, experience in
>>     GIN,  >
>>     >
>>     >> overview of existing systems, generalized system and suggested  >
>>     >> areas for standardization.
>>     >>
>>     >> Please have a look before the call today.
>>     >>
>>     >> Felix Ehm will lead the discussion.
>>     >>
>>     >> Thanks,
>>     >> ----
>>     >> Hiro Kishimoto
>>     >>
>>     >>
>>     >> --
>>     >>   ogsa-wg mailing list
>>     >>   ogsa-wg at ogf.org <mailto:ogsa-wg at ogf.org>
>>     >>   http://www.ogf.org/mailman/listinfo/ogsa-wg
>>     <http://www.ogf.org/mailman/listinfo/ogsa-wg>
>>     >>
>>     >>
>>     >
>>     > --
>>     >   ogsa-wg mailing list
>>     >   ogsa-wg at ogf.org <mailto:ogsa-wg at ogf.org>
>>     >   http://www.ogf.org/mailman/listinfo/ogsa-wg
>>     <http://www.ogf.org/mailman/listinfo/ogsa-wg>
>>     > _______________________________________________
>>     > glue-wg mailing list
>>     > glue-wg at ogf.org <mailto:glue-wg at ogf.org>
>>     > http://www.ogf.org/mailman/listinfo/glue-wg
>>     <http://www.ogf.org/mailman/listinfo/glue-wg>
>>
>>     _______________________________________________
>>     glue-wg mailing list
>>     glue-wg at ogf.org <mailto:glue-wg at ogf.org>
>>     http://www.ogf.org/mailman/listinfo/glue-wg
>>     <http://www.ogf.org/mailman/listinfo/glue-wg>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> glue-wg mailing list
>> glue-wg at ogf.org
>> http://www.ogf.org/mailman/listinfo/glue-wg
>>   
>>     
>
> --
>   ogsa-wg mailing list
>   ogsa-wg at ogf.org
>   http://www.ogf.org/mailman/listinfo/ogsa-wg
>   

-- 
http://www.cs.cf.ac.uk/User/O.F.Rana/index.html / work-fax:+44(0)29-2087-4598
work:+44(0)29-2087-5542 / other:+44(0)7956-299981 / distributed collaborative
computing / room n2.14 / school of computer science / cardiff university 
  queen's buildings / newport road / cardiff cf24 3aa / wales / uk 



More information about the ogsa-wg mailing list