Large Scale Jobs Reimbursement Program

From: Francesca Verdier (fverdier_at_lbl.gov)
Date: 12/04/2003


Greetings Seaborg users,

DOE and NERSC have set aside 1 million SP hours to encourage
projects to investigate code efficiencies on 64 and more nodes
(1,024 + processors) and to encourage running codes on these
large node configurations that focus on productive science progress.

The 1 million SP hours will be made available, under the 
conditions below.  

(1) If your project has codes that have already scaled to 1,024
processors the project is eligible to be reimbursed up to 100,000
SP hours per month for jobs run on 1,024 or more processors.  The
total reimbursement for such jobs is 750,000 hours.

(2) If your project has not yet scaled to 1,024 processors the
project is eligible to be reimbursed up to 100,000 SP hours per month 
for jobs run on 760 or more processors.  At least 250,000 hours
(of the 1 million currently available) is reserved for projects
in this category.

Reimbursements will be made towards the beginning of each month  
so long as you submit a paragraph to NERSC explaining the science 
results achieved with these large runs.  You must also gather
performance information for your large runs using the poe+
command; see:

http://hpcf.nersc.gov/software/ibm/hpmcount/

To request participation in the reimbursement program send a message
to NERSC Consulting explaining briefly what you hope to accomplish
by participating in the program:

Web:    http://help.nersc.gov
Email:  consult@nersc.gov

Sincerely,
-- 
Francesca Verdier                   email: fverdier@lbl.gov
Group Lead, NERSC User Services     phone: 510-486-7193

This archive was generated by hypermail 2.1.6 : 08/21/2008 PDT