[gin-info] gin-jobs use case

[Laurence Field]

This is a very good question!

I think that we only need to know the end point for the CE. I am sure 
that some people could argue for ever about what an endpoint was, but 
for me in this activity it just a string which I can use to contact the 
service and submit a job.

I think that the string you suggested should be fine. The only question 
I have is should we prefix this string with the protocol? eg

gram://hostname:2119/jobmanager

Laurence

Gerson Galang wrote:
> Hi Laurence,
>
> Can you tell me what type of Compute Resource information you need us to
> publish? Do want us to publish all the basic Globus services our site provides
> (eg GRAM, GridFTP, MDS, etc)? When you say endpoint, is it just in this format
> hostname:2119/jobmanager for the GRAM service?
>
> Thanks,
> Gerson
>
>
> Quoting Laurence Field <Laurence.Field at cern.ch>:
>
>   
>> At OGF 18, a new use case was proposed. gin-jobs would like to discover all
>> the Compute Resources which are accessible by the GIN VO.
>>
>> The core bits of information that need to be published for each service are.
>>
>> ServiceUniqueID
>> ServiceName
>> ServiceType
>> ServiceVersion
>> ServiceEndpoint
>> ServiceAccessControlRule
>>
>>
>> All this information can be found in the GlueService entry.
>>
>>
>> Shortly, I will try modifying the translators so that all the information is
>> reported correctly for this use case.
>>
>> Please let me know if you have any comments on this use case or the
>> attributes needed.
>>
>> Thanks
>>
>> Laurence
>>
>>
>>
>>
>> --
>>   gin-info mailing list
>>   gin-info at ogf.org
>>   http://www.ogf.org/mailman/listinfo/gin-info
>>
>>     
>
>