[GHPN-WG] Minutes from meeting @ OGF20 Manchester

[Cees de Laat]

Dear GHPN'ers,

Upfront a remark from myself. As most of you now know my term as 
AD-infrastructure ended and I indicated not to be available for 
another term because of mostly workload @ University of Amsterdam 
reasons. However, I indicated last year, when Dimitra was asked to 
become the sole chair, that I might wanted to to help her after my 
AD-term and so happens this week she twisted my arm.
Therefore, the process is to ask consensus from the RG and approval 
from the AD's. At the meeting there was no objection and the present 
AD RHJ gave the oke. And as you know decisions need to be verified on 
the mailinglist. So if anyone objects me being co-chair, please feel 
free to say so.

I took notes at the meeting, see below. Please inform me for 
additions, corrections, etc.

Best regards,
Cees.


Here are the rough notes from the GHPN meeting OGF20, 18h00-19h30 
wednesday may 9th.

agenda:

Dimitra Simeonidou(GHPN-RG)
Group Discussion GHPN
SessionAgenda:
1. Agenda bashing and administration (Dimitra Simeonidou) [5']
2. Update on GUNI (Georgios Zervas) [20]
3. Enlightened Computing Update: An HD-class example (Jon MacLaren)[15]
4. SIP protocol for Grid Networks (Aldo Campi) [15]
5. Research Challenges for Optical Grid Networks (Marc De Leenheer) [15]
6. Introducing EC-GIN: Europe-China Grid InterNetworking (Sven Hessler)  [20]

- agenda bashing, no changes

- Dimitria announced she asked previous AD Infrastructure Cees de 
Laat to become co-chair of this group. He said yes if group approves. 
Group ha no objection. Announce on mailinglist and ask AD to approve 
and submit to Joel.

- We welcome Richard Hughes Jones as AD-Infrastructure.

- update on GUNI by Zervas.
Q  What is the benefit of grid aware gmpls layer
A can react faster, can dynamically change path, milliseconds scale 
adaptation, net-awareness for optimization, dynamic allocation.
Q what grid projects support this
A Phosphorus, Enligtend, g-lambda

- Enlightend project , mc Laren
(I missed the Q/A from Richard Hughes Jones)

- SSIP talk by Aldo Campi
Q SIP - IMS vs SIP in GRID. Wy SIP rather than SOAP? Complexity of 
sip stack is big, same overhead as XML, SIP has not good security
Q why is sip beteer
A sip is proven to scale, implementations for many services

- Research Challenges for Optical Grid Networks, Mark de Leenheer
  Q what is relation with grid of hybrid netw (cirq plus obs)
      if you change internet in web app, what does that mean for grid apps.
A can remove complexity from the tcp stack

- 6. Introducing EC-GIN: Europe-China Grid InterNetworking (Sven 
Hessler) Q/A Please fill Questionaire from GIN. Put it on mailing 
list.
Q do not use tcp because of slow recovery
A are not going to use tcp. Suggestion from audience: Discuss with IRTF.

Meeting concluded on time.

Best regards,
Cees.
-- 
http://www.science.uva.nl/~delaat/