, view all tags
-- PeterKammel - 08 Apr 2008

* Persons/group who can change the list:

This page documents our input and documentation on our NCSA proposals for the MuX projects.


Proposal 2006

Proposal Duration Referee report Comments
prop06 4/1/06 to 3/31/07 see Peter partially approved on condor
response06a     full approval on tungsten
Extension 0?      
Extension1 to 3/30/08    
Extension2 to 6/30/08    

Proposal 2008



Proposal Duration Referee report Comments


Extension MuCap 1

PI Name: Peter Kammel Proposal Number: PHY060019 Email: pkammel@uiuc.edu Phone: (217) 333-5424 Organization: Univ of Illinois, Urbana Extension Length Requested: 6 months Reason: The NCSA allocation has been essential for the success of this project. A first grad. student finished his Ph.D. and the results were published very recently (V. A. Andreev et al. (MuCap Collaboration) PRL 99 (2007) 032002). A new generation of students is analyzing a much larger data set. Progress has been excellent. However, the students are also involved in extended experimental runs in Switzerland and are careful to optimize the code and learn from each previous production pass to obtain complete, internally consistent results at an unprecedented precision level. Thus we request an additional six month extension of our allocation in order to refine our analysis and to optimally use the precious allocated SRUs.

Extension MuCap 2

PI Name: Peter Kammel Proposal Number: PHY060019 Email: pkammel@uiuc.edu Phone: (217) 333-5424 Organization: Univ of Illinois, Urbana Extension Length Requested: 3 months Reason: We request an extension of our current MRAC allocation for three months, following the 6 months extension expiring March 31, 2008. The additional 3-months are essential to finish a first full analysis pass over the 2006-2007 experimental data from the MuCap experiment and will utilize a significant fraction of our current allocation. This analysis will be an important component of our renewal proposal which will be submitted before Apr. 15, 2008 to begin the new allocation in July 1, 2008. We are aware

that this 3-months extension is an exception and have discussed this with the NCSA allocation staff. It is critical to the success of our multi-year project, which heavily relies on continuing access to the NCSA resources and the more than 40 TB of data already transferred to MSS. As requested, a multi-year data storage request for the University of Illinois Muon Physics Group will be submitted separately.

The NCSA allocation has become a central and indispensable part of the MuCap experiment. A first grad. student finished his Ph.D. and the results were published (V. A. Andreev et al. (MuCap Collaboration) PRL 99 (2007) 032002).

Two additional graduate students as well as several more senior collaboration members have since joined the analysis effort at NSCA on the much larger data set. The main reason for the present request for an extension is that most of the analysis team was also involved in the final and successful data taking runs at the accelerator at PSI, Switzerland, lasting over a period of several months. All the data of 40 TB collected in 2006-2007 has been transferred to MSS and the focus during the next years will be on the analysis.

A brief update on the analysis and software effort of the project at NCSA is as follows. During our current allocation period, our efforts focused on upgrading our software to address the additional data that has been collected. We utilized 10775 SU to develop the code and analyze 15% of the data generated in 2006. This small scale analysis has been critical in guiding our code development to address the outstanding physics issues in the data. Our short term plan is to perform an additional small scale analysis within the next 3 weeks in order to provide feedback on our most recent code development. After ensuring that the changes enable us to study the underlying systematics and physics with a precision commensurate with the much higher statistical power available in the new final data, we will then perform a first full pass over all of the data within the next 2-4 months. With 18 TB from 2006 and 22 TB from 2007, a full pass requires approximately 50 kSU, and generates 20 ! TB of derived, tree-structured data. Several passes over the tree data are anticipated, each taking roughly one tenth of the CPU time required for the first pass over the raw data.

Statistics : ProdPass1Run10

This link contains a table of consumed SU for an early production pass

Topic attachments
I Attachment History Action Size Date Who Comment
Microsoft Word filedoc MuLan_NCSA_MRAC_Proposal.doc r1 manage 540.0 K 2008-04-08 - 19:59 PeterKammel  
Microsoft Word filedoc prop06.doc r1 manage 621.0 K 2008-04-08 - 19:57 PeterKammel  
Microsoft Word filedoc response06a.doc r1 manage 158.5 K 2008-04-08 - 19:57 PeterKammel  
Edit | Attach | Watch | Print version | History: r21 | r4 < r3 < r2 < r1 | Backlinks | Raw View | Raw edit | More topic actions...
Topic revision: r2 - 2008-04-17 - BrendanKiburg
  • Edit
  • Attach
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2021 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback