January 2006 - Lans-scidac - mailman3.cels.anl.gov

Feedback from Strayer
by Jorge More' 30 Jan '06

30 Jan '06

Some have been getting the following feedback. Let me know if you get a different feedback. It seems like we may ahve full feedback by tomorrow's meeting. Jorge The following is sent on behalf of Michael R. Strayer, SciDAC Director **************************************************************************** **************** Based on the information provided in your letter of intent submitted under Office of Science Notice (DE-FG02-06ER06-04) "Scientific Discovery through Advanced Computing", we noticed a match between the research you propose and SciDAC. . Therefore, we encourage you to submit a formal application under this Notice. Sincerely, Michael R. Strayer, SciDAC Director

1 0

Information re SciDAC proposals
by Jorge More' 23 Jan '06

23 Jan '06

Hi, You are all busy sending letters of intent, but after you get done with that, I would like to assemble a list of SciDAC planned proposals. This will give Jim an idea of how many budgets he will need to prepare for this call. Please send me the following information for each letter of intent: - Title of proposal - Area of proposal: SA, CET, SAP, or SI (science Institute)? - PI for the proposal - Institution of PI - MCS point of contact/PI - Total funding request - Funding request for MCS In order to minimize duplications, only the MCS point of contact/PI should send the information. I will assemble the information and share it with interested parties. Thanks! Jorge

4 5

Information from Walt Polansky
by Jorge More' 23 Jan '06

23 Jan '06

I think that we know all of this except that the submission should be in either txt or Word. LBL is doing the submissions in Word. Jorge ------- Forwarded Message The following tips appear to come from Sandia ... I think we know most of them already. > -------------------------------------------------------------------- > Comments from Walt Polansky: > > 1. Letters of intent can be either plain text (.txt) or > Microsoft Word (.doc) formats. Their only requirement is that > they NOT be PDF as this hinders their ability to cut and paste > information into their review database. > > 2. He also said that they must be easily readable (go no smaller > than 10pt). > > 3. Walt emphasized that they want all the information listed in > the call for each letter of intent since they plan to cut/paste > this information out of the letters and into their review > database. > > a. Title > b. Program area (e.g. Science Application (BER), Science > Application (FES) with Partnership (ASCR), > CET, Institute) > c. Principal investigator > d. All other senior personnel (anyone that will be listed > explicitly by name on a budget page) > e. Participating institutions > f. Organizational approach (how will the project be managed? > You should be convincing that > this will be a true collaborative effort and not a loose > assembly of different efforts.) > g. Projected funding request (list per institution and, when > applicable, denote breakdown between > science and enabling technologies (e.g. for SA/SAP)) > h. Summary/Abstract > > 4. It would seem that the primary purpose of the letter of > intent is to help them prepare for the full proposal review. > > 5. He did say that folks shouldn't worry too much about SA > versus SAP. When/if a proposal is funded, they will then figure > out on a case by case basis how much is to be funded by ASCR and > how much from a science office (BER, FES, NP, NNSA, etc). > > 6. There have been some questions posed to be regarding SA and > SAPs. In general they should be integrated into a single > proposal under almost all circumstances (see comment 5 above). > The main exception is when the SA is already funded (say by > BES). In that case, a "stand alone" SAP can be written although > care should be taken to make sure that a clear and compelling > partnership exists between the existing SA and the new > Partnership. > > 7. Remember to use the correct subject line for your email and > to CC appropriate POCs [for example a SA/SAP on computational > biology would CC John Houghton (BER) and Anil Dean (ASCR)] > > 8. Only 1 letter of intent should be sent by the LEAD > INSTITUTION (defined as the PI's institution). > ------- End of Forwarded Message

3 2

Re: your mail (fwd)
by Barry Smith 15 Jan '06

15 Jan '06

---------- Forwarded message ---------- Date: Sun, 15 Jan 2006 09:18:53 -0600 (CST) From: Rick Stevens <stevens(a)mcs.anl.gov> To: Barry Smith <bsmith(a)mcs.anl.gov> Cc: more(a)mcs.anl.gov, David Keyes <kd2112(a)columbia.edu>, Rusty Lusk <lusk(a)mcs.anl.gov> Subject: Re: your mail Here is my read. They dont know which applications will be funded. It is perhaps the case that PERC is adequately generic that the binding can happen after awards are made. TOPS is I think quite different. Its algorithmic strategies are pretty focused on a set of applications targets.. you change the targets and you change the foucs. Having said all of that.. I was just in DC this week and spent 3 hours with Strayer discussing a range of topics. My sense from that meeting is that there are certain applications groups that are more strategic in the near term to ASCR than others. In particular the nuclear energy and nuclear physics areas (including those areas of chemistry and materials of interest to NNSA) are on the up-tick. Fusion is probably ok. Climate is ok, but I detect little future growth there. BES is a pre-emergent area in the sense that they are not big partners in this year's SciDAC but Michael is planning another $20M in SciDAC funding next year which would be targeted at partnering with NE and with BES. All in all I would say that the best strategy would be to not go out of one's way to include explicit connections unless they are required by the science or strategy of the proposal. If one does do that.. keep an eye on what is likely to be relevant for the future as much as the past. (i.e. the nuclear comment above).. and finally.. Strayer would like things to get more international. So that might be something to keep in mind for the future. We spend a fair amount of time talking about how to grow a international component to SciDAC, in the next couple of years. I hope this helps.. On Sat, 14 Jan 2006, Barry Smith wrote: > > Rick, > > This email is surprising to me: INSTRUCTED EXPLICITLY ... SHOULD NOT > POINT TO SPECIFIC SciDAC-2 APPLICATION PROPOSALS. > > Five years ago we did point to specific app proposals and this time > (in TOPS) I think we planned to even more, with technical discussion > of the specific algorithmic (math) problem that group needed to solve. > > (That is not just saying "we will work with XYZ", but saying "we will > work with XYZ on their bottleneck problem, computation of 10,000 eigenvalues > of 100,000 by 100,000 sparse matrices that arise in their code ..... We will > use and further develop algorithm XYZ that ....") > > Do you have any information on this EXPLICIT INSTRUCTION? > > Thanks > Barry > > > ---------- Forwarded message ---------- > Date: Sat, 14 Jan 2006 00:24:20 -0800 > From: Esmond G. Ng <EGNg(a)lbl.gov> > > In our proposal (not the LOI), are we going to refer to the general categories? > Or are we going to point to specific SciDAC-2 application proposals? > > I had a conversation with Bob Lucas about his PERC proposal. He said that he > was instructed explicitly by Fred Johnson that the PERC proposal should not > point to specific SciDAC-2 application proposals. However, exchange of support > letters would be OK. Perhaps we should check with the Math CET program manager > about this ;-) > > Cheers, > Esmond > >

1 0

info about letters of intent for SAPs
by Lois Curfman McInnes 13 Jan '06

13 Jan '06

FYI. ---------- Forwarded message ---------- Date: Fri, 13 Jan 2006 13:43:59 -0500 From: bernholdtde(a)ornl.gov To: cca2-proposal(a)cca-forum.org Cc: bernholdtde(a)ornl.gov Subject: [CCA2-Proposal] LOIs for SAPs >From a phone conversation with Fred Johnson: Where there is an application proposal going in simultaneously, a separate LOI for and SAPs is neither required nor desired. Where the target application is already established, obviously an LOI _is_ going to be required for the SAP. SAPs are always associated with specific application(s), be they existing ones or new proposals. In the RFP's instructions for emailing the LOIs, SAPs should be classed with applications, not with enabling technologies (this pertains to the info requested in the Subject line of submission emails). [This makes complete sense, but this is the first confirmation I've seen from a program manager of the basic expectations for LOIs. As Fred said, it is not mean to be tricky, just flexible.] -- David E. Bernholdt | Email: bernholdtde(a)ornl.gov Oak Ridge National Laboratory | Phone: +1 (865) 574 3147 http://www.csm.ornl.gov/~bernhold/ | Fax: +1 (865) 576 5491 _______________________________________________ CCA2-Proposal mailing list CCA2-Proposal(a)cca-forum.org http://www.cca-forum.org/mailman/listinfo/cca2-proposal

1 0

SciDAC-2 update
by Jorge More' 12 Jan '06

12 Jan '06

FYI, we are getting conflicting information on the format of the LOI. PDF may be acceptable or maybe not. Jorge ------- Forwarded Message - -------- Original Message -------- Subject: [SciDAC_PI] SciDAC-2 Update Date: Thu, 12 Jan 2006 11:08:52 -0800 From: Juan Meza <JCMeza(a)lbl.gov> To: scidac_pi(a)hpcrdm.lbl.gov All, Here's some new information on the SciDAC call in answer to some of your questions. 1) Institutes guidelines: There were some questions regarding budget splits between labs and universities for the Institutes. The guidance we received is that DOE has no preconceived notions regarding what proportion of the budget should go to either the labs or the universities, i.e. the labs can play a substantial role in the Institutes. 2) Letter of Intent: The guidance in the call is in fact correct regarding "*n**ot in PDF*". Either a Word document or plain text is needed. As I mentioned earlier, the SciDAC admin team is preparing a template which you can use. 3) Fonts and margins: The lab and universities do indeed have different guidelines. I highly recommend using the 11 point and 1 inch margin guidelines. 4) Use a starting date of July 1. 5) SAPs: The general guidance in the call was written to give the SAPs as much flexibility as possible. They can be activities within a Science Application (SA) proposal, they can be separate proposals, or anything in between. If they are wholly contained in an SA then the SAP doesn't need a separate letter of intent. If you're thinking about a self-contained SAP proposal then it needs its own letter of intent. 6) SAPS (cont.): Note that a SAP can constitute a partnership with an "*application science **area*" and not just a single application/code, i.e. a SAP can span several applications as long as you can make a case for it. 7) You may use one signed cover page from the lead institution for the final proposal. The other institutions must still send in their proposals through their normal channels, but the lead institutions don't need to collect all of the signed cover pages from everybody before submitting the proposal. If you have other questions, please let me know. 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> ------- End of Forwarded Message

1 0

Letters of Intent
by Paul Hovland 11 Jan '06

11 Jan '06

The SciDAC 2 call says: >Letters-of-Intent should be submitted electronically as an email >attachment (not pdf) to scidac-2(a)mics.doe.gov. > Does anyone know what "not pdf" is supposed to mean? Do they want plain text? Word? LaTeX ;-)? Paul

2 1

Fwd: Re: SciDAC 2 Call - Budget Information (fwd)
by Barry Smith 11 Jan '06

11 Jan '06

---------- Forwarded message ---------- Date: Wed, 11 Jan 2006 10:23:51 -0600 From: Ian Foster <foster(a)mcs.anl.gov> To: big-sammy(a)mcs.anl.gov, dsl-management(a)mcs.anl.gov Subject: Fwd: Re: SciDAC 2 Call - Budget Information >Date: Wed, 11 Jan 2006 08:05:45 -0800 >From: "Dean N. Williams" <williams13(a)llnl.gov> >User-Agent: Mozilla Thunderbird 1.0.7 (Macintosh/20050923) >X-Accept-Language: en-us, en >To: Don Middleton <don(a)ucar.edu>, bernholdtde(a)ornl.gov, > Ian Foster <foster(a)mcs.anl.gov> >Subject: Re: SciDAC 2 Call - Budget Information >X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at mailgw.mcs.anl.gov > >======================================================= >Here are the highlights from our discussion with Walt Polansky. > > * *Total funding available for SciDAC-2. *OASCR's total contribution > to the SciDAC-2 call is approximately $40M. The total amount of > funding available in the SciDAC program is, in fact, around $66M. > (note: based on our information, it appears that somewhere between > $50M and $60M is available for this call; the difference is in > ongoing application activities not being recompeted this year) > * *Application proposals. *When asked about the notion of submitting > large integrated proposals in the applications areas rather than a > number of smaller proposals, Walt responded that the general > spirit of the SciDAC program is to create large, critical mass > efforts in each of the areas highlighted in the call. Walt has a > preference to see, for example, 2 large proposals in a given area > rather than 10 small proposals. But confirmation from the > application program managers should be sought through the > appropriate laboratory point of contact > * *Start dates; budget pages. *The start date for these proposals > should be July 1. Walt would prefer to see the budget sheets as > full years starting in July. It may be that the FWP requires > fiscal year input, and we will check on that. > * *Discussion of SAP proposal options*: > o OASCR's SAP funding will most likely be used only to fund > mathematicians and computer scientists working on SAP > activities; it will not fund applications scientists. > Funding for application scientists on stand-alone SAP > proposals (see below) will be provided by the application > program office. > o Applications may propose *MORE* than $500K in SAP activities > and should propose what makes sense; the guidelines here (as > elsewhere in the call) are just that, guidelines. In > general, the SAP activities should be discussed in the main > body of the proposal; the appendix should not be used for a > technical description of SAP activities. > o See attached PowerPoint slide where: app folks are red, app > activities pink, Math/CS folks are blue, Math/CS activities > light blue. SAP activities are green. CET activities are cyan. > > _______________________________________________________________ Ian Foster www.mcs.anl.gov/~foster Math & Computer Science Div. Dept of Computer Science Argonne National Laboratory The University of Chicago Argonne, IL 60439, U.S.A. Chicago, IL 60637, U.S.A. Tel: 630 252 4619 Fax: 630 252 1997 Globus Alliance, www.globus.org

1 0

How climate is handling SAPs
by Jorge More' 10 Jan '06

10 Jan '06

Just FYI: > > > > >I just found out that there will be about 10 or so SAP proposals > >affiliated with the climate application area. The climate application > >folks will not endorse any of the proposals. They're just going to add the > >SAPs as appendices to their proposal. > > > >-Dean >

2 1

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