- Lans-scidac - mailman3.cels.anl.gov

SAP proposals
by Jorge More' 10 Jan '06

10 Jan '06

Rick, There is a need to clarify the SAP proposal submission. Below is a note from Pat Worley re SAPs that Paul Hovland circulated. My reading is that if you submit a SAP, then you must submit a letter of intent and a proposal. This would be the process in all cases, not as Pat mentions, for multiple institutions or if led by an institution different from the app. However, in all cases, as Pat says, > (SAP still needs to be called out in App proposal and > budget included in App appendix)." Is this the process? Jorge ------- Forwarded Message From Pat Worley (ORNL): > IMPORTANT: the latest information that I have, and it seems to change > daily, is that > > "if a SAP has multiple institutions, or is led by an institution > different than the lead App institution, then > the SAP must submit a letter of intent and separate proposal. > (SAP still needs to be called out in App proposal and > budget included in App appendix)." > > so if you are thinking about a SAP you need to get buy-in from the > application > lead and prepare a letter of intent ASAP. ------- End of Forwarded Message

2 1

Some info on SAPs
by Paul Hovland 09 Jan '06

09 Jan '06

From Pat Worley (ORNL): > IMPORTANT: the latest information that I have, and it seems to change > daily, is that > > "if a SAP has multiple institutions, or is led by an institution > different than the lead App institution, then > the SAP must submit a letter of intent and separate proposal. > (SAP still needs to be called out in App proposal and > budget included in App appendix)." > > so if you are thinking about a SAP you need to get buy-in from the > application > lead and prepare a letter of intent ASAP. Not sure what "led by" means in the case of a SAP. Paul

1 0

FAQs posted for SciDAC-2 Call (fwd)
by Barry Smith 07 Jan '06

07 Jan '06

---------- Forwarded message ---------- Date: Sat, 7 Jan 2006 07:52:30 -0600 From: "Bair, Raymond A." <rbair(a)anlmail.anl.gov> To: big-sammy(a)mcs.anl.gov Subject: FAQs posted for SciDAC-2 Call See these URLs for FAQs related to SciDAC-2: University call: http://www.mics.doe.gov/scidac/scidac2univfaq.html Lab call: http://www.mics.doe.gov/scidac/scidac2labfaq.html Ray

1 0

Ian's information
by Jorge More' 05 Jan '06

05 Jan '06

This is oldish informationfrom Ian, but it may be useful. Jorge ------- Forwarded Message From: Ian Foster <foster(a)mcs.anl.gov> Subject: SciDAC-2 Proposals I wanted to mention what I know about SciDAC-2 proposals that are in the works. Hopefully this will allow us all to identify opportunities for collaboration, gaps, etc. * Distributed Systems ISIC or Institute: I am putting this together with, so far, LBNL as the main lab partner and a number of university partners. The focus is probably going to be on distributed data management, authorization, and perhaps workflow. HELP: I'd really appreciate input from anyone involved in a SciDAC science project concerning requirements that you perceive in this space. * Security ISIC: Deb Agarwal of LBNL is putting this together, with so far ANL and PNNL as the main lab partners. The focus is on the operational security issues that must be addressed to allow resource access and sharing. Frank Siebenlist and I are currently the Argonne representatives. * Climate data analysis SAP: This is a follow on to the Earth System Grid project, focused on enabling access to, and analysis of, the increasingly large quantities of data being produced by climate models, e.g., for the next IPCC assessment. I am currently the Argonne representative. * Nanosciences SAP: Gregor von Laszewski is working with Derrick Mancini of CNM to develop ideas for a project that would span multiple nanoscience centers. I don't know more about this yet. * Open Science Grid: The Open Science Grid consortium is developing a proposal to be submitted to DOE and NSF to operate and extend the Open Science Grid infrastructure that was developed under PPDG/GriPhyN funding. I am currently the Argonne representative. * Fusion collaboratory: This community would seem a natural one to write a proposal, given ITER. However, they are holding off until they see the call. ------- End of Forwarded Message

1 0

[Fwd: more SciDAC2 info]
by Paul Hovland 05 Jan '06

05 Jan '06

This includes Al Geist's full message, not just the first bullet that Barry forwarded earlier. Paul

1 0

Planning meeting
by Jorge More' 05 Jan '06

05 Jan '06

Hi, Now that we have some basic information on SciDAC, I would like to have a short meeting to share information and to plan for the Jan 23 letters of intent. The main items are - Which letters of intent are we submitting? - What other issues does DOE need to clear up re the call? I am mostly concerned with LANS participation, but interested parties are invited. The meeting will be this Friday at 1pm in A261. Jorge

1 0

finally a no-nonsense answer (not surprisingly from Fred)
by Barry Smith 04 Jan '06

04 Jan '06

-----Original Message----- From: Al Geist [mailto:[email protected]] Sent: Wednesday, January 04, 2006 4:35 PM To: Samatova, Nagiza F.; Scott, Stephen L.; Bernholdt, David E.; Jack Dongarra; Nichols, Jeffrey A. Cc: Geist, Al; Zacharia, Thomas Subject: SciDAC $300K per institute answer from MICS Hi SciDAC PIs, Feel free to pass the following on to others who need to know. Since each of you asked me this question, I've included you on the initial reply. I talked to Fred Johnson in MICS and here are his answers to my questions: 1. The guidance says there is a $300K per institute limit for CSETs and SAPS. What's up with that? Fred: This is the #1 question I have been getting. the $300K limit was put into the university call and I suspect didn't get edited out of the Lab call, but the limit is not reasonable for Labs and they should put in for amounts much like they did for SciDAC 1.

1 0

Re: [Fwd: SciDAC Meeting Summary] (fwd)
by Barry Smith 04 Jan '06

04 Jan '06

(All for $300,000 per site, gosh ASCAR is a mess). ---------- Forwarded message ---------- Date: Wed, 04 Jan 2006 16:40:18 -0600 From: Boyana Norris <norris(a)mcs.anl.gov> To: Barry Smith <bsmith(a)mcs.anl.gov> Cc: pieper(a)mcs.anl.gov Subject: Re: [Fwd: SciDAC Meeting Summary] (fwd) I am wondering about this point: "6) The SciDAC institutes are defined very broadly, but should be modeled along the lines of the UCSB ITP and the Santa Fe Institute." I don't know much about the ITP and the Santa Fe institutes, but from browsing their pages (at least what I think are their pages), it seems that they are significantly larger (dozens of people) and much less temporary institutions. Does anyone have a good idea what "modeled after" means in this case? Perhaps I'm looking at the wrong pages, so I'd appreaciate if someone can send links. Boyana P.S. Barry, can you please forward this to lans-scidac? For some reason I get a mail server error when I try to send to the list. Barry Smith wrote: > FYI, > > > > 1) The $3M and $300K/institutions limits are for guidance only. Proposals > > will still be judged on the technical quality and funding decisions could > > be made that will exceed these limits, if the technical merit of the > > proposal supports higher funding. > > What a crap answer; this is exactly what the call already says. > > And where do some people at some labs get off having privating meetings > with the people who end up making the funding decisions? > > Barry > > ---------- Forwarded message ---------- > Date: Wed, 04 Jan 2006 14:10:37 -0800 > From: Esmond G. Ng <EGNg(a)lbl.gov> > To: David Keyes <kd2112(a)columbia.edu>, Barry Smith <bsmith(a)mcs.anl.gov>, > Jorge More' <more(a)mcs.anl.gov>, Rob Falgout <rfalgout(a)llnl.gov>, > Carol Woodward <cswoodward(a)llnl.gov>, Xiaoye S. Li <XSLi(a)lbl.gov> > Subject: [Fwd: SciDAC Meeting Summary] > > As of possible interest ... > > -------- Original Message -------- > Subject: SciDAC Meeting Summary > Date: Wed, 04 Jan 2006 13:50:22 -0800 > From: Juan Meza <meza(a)hpcrdm.lbl.gov> > To: HPCRD GL <grpleads(a)hpcrd.lbl.gov>, Horst Simon <HDSimon(a)lbl.gov>, > "Michael J. Banda" <MJBanda(a)lbl.gov>, David H Bailey <DHBailey(a)lbl.gov>, Deb > Agarwal <DAAgarwal(a)lbl.gov>, Victor M Markowitz <VMMarkowitz(a)lbl.gov> > CC: scidac-admin <scidac_admin(a)hpcrd.lbl.gov> > > > > All, > > Horst and I met with Michael Strayer and Walt Polansky today to clarify some > of > the questions we've had regarding the SciDAC-2 call. Here's a short summary > of > my notes. I will be back in town on Friday and can discuss these further with > you or you can send me email if you have specific questions. > > > 1) The $3M and $300K/institutions limits are for guidance only. Proposals > will > still be judged on the technical quality and funding decisions could be made > that will exceed these limits, if the technical merit of the proposal supports > higher funding. > > 2) The stakeholder apps in proposals can extend beyond the ones in the call. > However, PIs must ensure that there is DOE mission relevance. > > > > 3) The software lifecycle should be clearly addressed in the proposal. This > has > taken on a stronger emphasis than in SciDAC-1. > > 4) A clear path/process for how the software tools will be made open source > should be described. > > 5) The letters of intent should contain an organizational approach section > that > describes the relationship between all of the PIs and the participating > institutions. > > 6) The SciDAC institutes are defined very broadly, but should be modeled along > the lines of the UCSB ITP and the Santa Fe Institute. > > 7) Education and outreach will play a part in the evaluation (perhaps mostly > in > the Institutes portions?). Graduate students are considered part of the > education process, whereas postdocs are not counted. > > 8) The letters of intent as well as the proposals must have 1 > lead/coordinating > PI. > > 9) SciDAC-2 proposals can have a research component. > > 10) There will be a FAQ page on the DOE web site soon. Check this FAQ page > for > answers to the following questions: > a) Will labs be able to compete for the NNSA projects and funding? > b) Is it possible to submit a SAP proposal that links to an already funded > project, e.g. BES or FES SciDAC projects that were started or recompeted in > 2004 > and 2005? > c) Clarification of the limit of 5 pages per task. > > > Best, > > Juan > > */Department Head/* > > */High Performance Computing Research/* > > */Lawrence Berkeley National Laboratory/* > > */One Cyclotron Road, MS: 50B-2239/* > > */Berkeley, CA 94720/* > > */http://hpcrd.lbl.gov/~meza <http://hpcrd.lbl.gov/%7Emeza>/* > > > -- "I don't think anyone should write their autobiography until after they're dead." Samuel Goldwyn (1882 - 1974)

1 0

[Fwd: SciDAC Meeting Summary] (fwd)
by Barry Smith 04 Jan '06

04 Jan '06

FYI, > 1) The $3M and $300K/institutions limits are for guidance only. Proposals will > still be judged on the technical quality and funding decisions could be made > that will exceed these limits, if the technical merit of the proposal > supports higher funding. What a crap answer; this is exactly what the call already says. And where do some people at some labs get off having privating meetings with the people who end up making the funding decisions? Barry ---------- Forwarded message ---------- Date: Wed, 04 Jan 2006 14:10:37 -0800 From: Esmond G. Ng <EGNg(a)lbl.gov> To: David Keyes <kd2112(a)columbia.edu>, Barry Smith <bsmith(a)mcs.anl.gov>, Jorge More' <more(a)mcs.anl.gov>, Rob Falgout <rfalgout(a)llnl.gov>, Carol Woodward <cswoodward(a)llnl.gov>, Xiaoye S. Li <XSLi(a)lbl.gov> Subject: [Fwd: SciDAC Meeting Summary] As of possible interest ... -------- Original Message -------- Subject: SciDAC Meeting Summary Date: Wed, 04 Jan 2006 13:50:22 -0800 From: Juan Meza <meza(a)hpcrdm.lbl.gov> To: HPCRD GL <grpleads(a)hpcrd.lbl.gov>, Horst Simon <HDSimon(a)lbl.gov>, "Michael J. Banda" <MJBanda(a)lbl.gov>, David H Bailey <DHBailey(a)lbl.gov>, Deb Agarwal <DAAgarwal(a)lbl.gov>, Victor M Markowitz <VMMarkowitz(a)lbl.gov> CC: scidac-admin <scidac_admin(a)hpcrd.lbl.gov> All, Horst and I met with Michael Strayer and Walt Polansky today to clarify some of the questions we've had regarding the SciDAC-2 call. Here's a short summary of my notes. I will be back in town on Friday and can discuss these further with you or you can send me email if you have specific questions. 1) The $3M and $300K/institutions limits are for guidance only. Proposals will still be judged on the technical quality and funding decisions could be made that will exceed these limits, if the technical merit of the proposal supports higher funding. 2) The stakeholder apps in proposals can extend beyond the ones in the call. However, PIs must ensure that there is DOE mission relevance. 3) The software lifecycle should be clearly addressed in the proposal. This has taken on a stronger emphasis than in SciDAC-1. 4) A clear path/process for how the software tools will be made open source should be described. 5) The letters of intent should contain an organizational approach section that describes the relationship between all of the PIs and the participating institutions. 6) The SciDAC institutes are defined very broadly, but should be modeled along the lines of the UCSB ITP and the Santa Fe Institute. 7) Education and outreach will play a part in the evaluation (perhaps mostly in the Institutes portions?). Graduate students are considered part of the education process, whereas postdocs are not counted. 8) The letters of intent as well as the proposals must have 1 lead/coordinating PI. 9) SciDAC-2 proposals can have a research component. 10) There will be a FAQ page on the DOE web site soon. Check this FAQ page for answers to the following questions: a) Will labs be able to compete for the NNSA projects and funding? b) Is it possible to submit a SAP proposal that links to an already funded project, e.g. BES or FES SciDAC projects that were started or recompeted in 2004 and 2005? c) Clarification of the limit of 5 pages per task. Best, Juan */Department Head/* */High Performance Computing Research/* */Lawrence Berkeley National Laboratory/* */One Cyclotron Road, MS: 50B-2239/* */Berkeley, CA 94720/* */http://hpcrd.lbl.gov/~meza <http://hpcrd.lbl.gov/%7Emeza>/*

1 0