perfsonar development work

[Andreas Hanemann <>]

Hi Jeff,

You are right. The implementation of the service will be different, but
the interface for the perfSONAR interactions will stay the same.

For the cNIS installation the aim is that it is not more complicated
than the TopS installation and that you can choose what kind of
information you want to provide with it. So that when you install the
cNIS you also have the option to use it for perfSONAR purposes only. The
TopS development should stop at some point to avoid a parallel work. In
retrospective this is not really optimal, but the need for the
cross-activity cNIS had not been so obvious in Y2 so that only came to
the agenda of Y3.

Best regards
Andreas


Jeff W. Boote wrote:
> Andreas Hanemann wrote:
>> Yes, you are right. The cNIS service will replace the TopS and it will
>> cover requirements from (at least) JRA1, JRA3, JRA4, and SA3.
>
> I don't disagree, but I would like to clear one issue up. The TopS
> perfSONAR interface/protocol is not being replaced. I realize when you
> say replace, you probably mean replace the current TopS implementation.
>
> I only bring this up to make sure we distinguish between the TopS
> "interface/protocol" and the TopS implementation. I think these two
> things could be getting confused in this discussion. (If not, please
> ignore this message. ;) )
>
> From an implementation perspective in JRA-1, there is no problem with
> any of the discussions I have seen. But I bring this up because of
> interoperability concerns.
>
> I would say that cNIS needs to be an implementation of the perfSONAR
> TopS service specification. (cNIS can of course implement more
> interfaces.) But, to ensure interoperability with all the other
> networks in the world that are unlikely to deploy a full cNIS, we
> (perfSONAR) need to define very well the perfSONAR TopS interface and
> protocols.
>
> jeff

-- 
Andreas Hanemann, 

Boltzmannstrasse 1, 85748 Garching, Germany
Telefon:   +49 89 35831-8712
Fax:       +49 89 35831-9700