perfsonar development work

[Guilherme Fernandes <>]

Loukik Kudarimoti wrote:

> Jeff W. Boote wrote:
>
> > Loukik Kudarimoti wrote:
> >
> > > Hi,
> > >
> > > While testing out RRD MA (from perfsonar 1.0) and SQL MA (snapshot), 
> > > I noticed that the response messages for a SetupDataRequest *without 
> > > key* message does not provide a key back along with measurement data 
> > > in the response. I remember that this wasn't the case some time ago.
> > >
> > > I support the idea of sending a key back for such requests. It 
> > > allows us to say "Here is the matching metadata and data for your 
> > > request and also, here is a key that you can use the next time you 
> > > want data for the same metadata".
> > >
> > > Based on my experience, I believe including the key in the response 
> > > requires a very minor change in the service engine code. All 
> > > services implementing the key mechanism should similarly provide the 
> > > key back for such requests.
> > >
> > > Any comments?
> >
> >
> > This sounds reasonable to me.
> >
> > I would also add that the metadataid should be stable and unique. I 
> > asked developers about a month ago how difficult that would be, and 
> > have yet to hear a single response.
>
> I can think of a few questions related to this topic.
>
> a) What is the significance of having unique ids?
> I am assuming that the ids need to be globally unique. For me, there 
> is no question about ids having to be unique in a message or a store. 
> Since I am not clear about the motive behind unique ids, I would 
> appreciate an answer to this question.
At least for the SQL MA: the 'primary key' (although the field is not 
configured that way, but I think it should for performance issues) of 
the table that stores data in the database is the field 'metadataId'. 
This is the only thing that relates data stored to a specific 'key' (as 
in eXist keys).

> b) What is the recommended structure of these ids (which one should 
> follow in order to generate a unique id)? This question is important 
> because clients such as viz. tools will need to generate these unique 
> ids.

In the Owamp cenario we created here at RNP, we make unique ids 
appending the timestamp of the time of the request to the metadata-id 
(done by the CLMP when he receives the request so it can keep track of 
it, it could also be done by the client, but I guess that not by the 
MA). In the case of RRD MA and SQL MA related to utilization I dont know 
how this would be done.

> b) Is it necessary to confirm that the id is unique? If so, how do we 
> go about doing that (rather, how can a program go about doing that)?

This could easily be done by a query to the eXist database. Also, the 
id's only have to be unique for that specific MA (I guess this was 
already being considered hehe).

That's from a CLMP -> MA perspective.

> c) In the scenario where client X sends a request with certain 
> metadata and data elements to Service A and Service A responds with 
> metadata and data elements to client X
> (i) If the service adds new metadata elements, the service generates 
> the ids for these elements?
> (ii) If the service modifies existing metadata or data elements, is it 
> required to modify the ids as well?
>
> d) Is it necessary to use unique ids for contents inside a store (i.e. 
> metadata configuration files)?
>
> Loukik.
>
> > Silence equals agreement.
> >
> > jeff