From: Albert Lazzarini <lazz@ligo.caltech.edu>
Date: Tue Jan 15, 2002  10:29:00  US/Pacific
To: LSCCompComm@ligo.caltech.edu
Cc: Albert Lazzarini <lazz@ligo.caltech.edu>, "Gary H. Sanders" 
<sanders@ligo.caltech.edu>, Rainer Weiss <weiss@ligo.mit.edu>, Barry Barish 
<barish@ligo.caltech.edu>
Subject: Minutes of meeting of 2002.01.15

Kickoff meeting for the LSC Computing Committee. The agenda and minutes (taken by 
A. Lazzairin) were as follows:

>1.	Committee charter -- discussion
>2.	Frequency of meetings, format, reporting to LIGO and LSC management, etc.
>3.	Assignment of priorities among:

>-	LSC and Grid computing activities in GriPhyN and iVDGL: requirements for LIGO 
computing
>-	LIGO-LSC MOUs for institutions that have _significant_ computing resources; 
how should LIGO lab enter into agreement with institutions regarding services and 
functions to be provided to the LSC by such institutions?
>-	Data Analysis White Paper - resource allocation mechanism, use review 
mechanism, etc.
>-	LSC interaction and participation in international grid computing 
collaborative activities -- GEO and Virgo.

4.	Any other order of business?


Attendees: W. Anderson, J. K. Blackburn, P. Brady, L.S. Finn, E. Katsavounidis, A. 
Lazzarini, T. Nash, R. Weiss, A. Wiseman

1.	The committee discussed the charter and the consensus was that it was 
appropriate for the time being, although this may need to be revisited this as 
time passes.

2.	The committee agreed on meeting every other week.  Albert will ask Rita 
Torres to poll the committee members in order to identify an acceptable 1 hour 
period during the week when people can meet. It was also agreed that the committee 
would convene whenever a quorum of 5 members (NOTE: Albert interprets this as 5 
INCLUDING the chair: please let him know if this is not the intended 
interpretation) could meet. Albert will ask one of the other committee members to 
take minutes at each meeting. These will be circulated to the LSC Spokesman, the 
LIGO Laboratory Directorship, and the committee. Patrick volunteered to assist 
Albert in setting up mailman to provide an email archive for the committee 
correspondence. The committee chair will report at the monthly LSC executive 
committee and also at the semi-annual LSC meetings in the same manner as the the 
software committee reports.

3.	There was general discussion on how to set priorities for the issues before 
this committee. The issues discussed were as follows.

-	The need to define a template for a new attachment to the LIGO MOUs with LSC 
member institutions for those institutions with SHARABLE resources (i.e., Tier 2 
or higher levels of resources). The new attachment would address commitments to be 
made to the LSC and LIGO to provide access to resources. How these resources would 
be used and how access is provided needs to be defined. There was discussion 
whether this committee should be doing that or whether another mechanism should be 
found,. It was agreed that this committee should take care of adjudication of 
resources as a standing agenda item to be discussed at every meeting. The need was 
also identified to reconcile past undestanding with future expectations regarding 
what Tier 2 centers are expected to provide to the collaboration as a whole.

-	There was a discussion how the yet-to-be defined LIGO data flow model would 
affect resource allocation and resource activities and responsibilities of the 
Tier 2 centers. Nash pointed out that the LSC has insufficient experience to date 
to be able to say what that model is; but that accruing experience will point the 
way in this regard. The committee needs to be prepared to track the evolution of 
this model and to adjust accordingly.

-	There was a discussion on how to use lessons-learned from the engineering 
runs and overall software development of the past year to factor the derived 
wisdom into what comes next with regard to developing grid-enabled applications 
and versions of LIGO search codes beyond the LAL/LDAS model. This activity would 
also lead to developing a refined definition of what will be needed in the next 
generation data analysis systems to support future LIGO upgrades (Advanced LIGO, 
e.g.)

-	It was proposed that this committee also compile and maintain a list of LSC 
resources that are governed by the MOUs. Lazzarini will develop a template for a 
form to be filled out by Tier 1 and Tier 2 centers that identify the resources. He 
will circulate this to the committee.

-	There was discussion on how future Tier 2 centers will be chosen as the 
collaboration's needs grow and as further opportunities to obtain funds for 
computing present themselves. The consensus was that this committee should develop 
those criteria by which LIGO directorship and the LSC management would be able to 
endorse future Tier 2 center proposers. The needd was also identified to provide 
guidance to new institutions regarding how to set up their resources in a mnaaer 
consistent with LSC expectations of Tier 2 centers.

-	There was a discussion on what the LSC participation in GriPhyN and iVDGL 
projects implies regarding commitments made to these other projects that need to 
be honored by the participating institutions. Finn suggested posting and 
circulating the iVDGL management and organization plan in order to inform this 
committee on these issues. It was also agreed that it would be desirable to invite 
LSC members who are participating in grid computing initiatives  (e.g., Scott 
Koranda) to make a presentation to this committee regarding Grid computing and LSC 
participation and what the expectations are from the grid projects on LSC 
acitivities to support the larger international activities. Related to this is the 
need to define how LIGO Tier 2 centers will be integrated with and contribute to 
the larger grid resource pool.

In concluding, the general ordering of priorities were as follows:

Highest: 	Develop the template for the new MOU attachment that will govern Tier 2 
resources and their usage.
High:	Invite LSC grid personnel to brief this committee on these activities and the 
integration issues discussed above.
High:	Develop and define the data flow model for LIGO data and the implications ofr 
this model on the use and activities of Tier 2 centers within the LSC.
High:	Develop a list of existing LSC computing resources and their availabilities
High:	Develop a grid usage and computing requirements document for the LSC -- 
follows the lead provided by the corresponding CMS document

Actions:

1.	Lazzarini to work with Torres to identify a suitable 1 hour period every 
other week for this meeting
2.	Lazzarini to develop and circulate a template form in order to determine what 
LSC resources are deployed and available across the collaboration
3.	Lazzarini to post iVDGL management and organizational plan to our wed site.

Please let me know if I have mis-stated or omitted anything.

Albert
-------------------------------------
LIGO Laboratory at Caltech       Data and Computing
M/C 18-34		                          Room 609 Millikan Library
1200 E. California Blvd.               Pasadena, CA 91125
Office: +1.626.395.8444               Facsimile: +1.626.304.9834
-------------------------------------