All NERSC Cori and Edison users are eligible to apply for early access to the regular KNL partition, which allows running jobs up to full system scale free of charge from March 1 to July 1. All users can currently run small jobs, but early access will enable users to run at scale.
To apply for early access, users will need to collect performance data on their application and fill out the application form at https://my.nersc.gov/knleap.php . We are accepting one application per repo, and the application can be submitted by any member of the repo. Please direct any questions to the consultants via my.nersc.gov, help.nersc.gov, or firstname.lastname@example.org .
Argonne’s intensive two-week summer training program on extreme-scale computing will be held July 30—August 11, 2017. The deadline to apply is March 10, 2017. Doctoral students, postdocs, and scientists interested in conducting computational science and engineering research on large-scale computers are encouraged to apply. For more information, visit: http://extremecomputingtraining.anl.gov/
Both CMS and ATLAS are amongst the six projects selected for the program that offers NERSC and vendor support in porting codes to the KNL architecture.
NERSC will host a three-day training event for users on Cori KNL Programming Environment, Debugging and Optimization from Tuesday February 14 to Thursday February 16. The main presenters will be from Cray.
Workshop for Beyond Leading Order Calculations on HPCs
Hosted by Argonne & FermiLab LPC
Sponsored by the DOE HEP Center for Computing Excellence
Coordinators: Taylor Childers & Liz Sexton-Kennedy
Time: September 22-23 (9am – 6pm on Thursday, 9am – 12pm on Friday)
Location: Fermilab, specific room to be announced.
Indico Agenda Link
The DOE will spend $0.5B over the next three years to replace current supercomputing resources with new machines. Current machines provide about 15B core-hours per year to scientific research, with LHC experiments using around 100M core-hours per year. The new machines will provide more than an order of magnitude increase in the computing resources available at the HPC institutes such as the NERSC (National Energy Research Scientific Computing) center, the Argonne Leadership Computing Facility, and the Oak Ridge Computing Facility. HEP must focus on using these machines effectively in order to achieve the science goals set out by the P5 report. This workshop brings together developers of parton event generators and software experts from HPC institutes in order to work toward HPC friendly codes that allow for the most compute intensive calculations to scale to millions of parallel processes.
NERSC will host a four-day training event for Cori Phase I users on programming environment, debugging, and optimization, from Monday to Thursday, June 13-16, 2016 at Wang Hall/CRT on the Berkeley Lab campus. The course, taught by Cray instructor Rick Slick, will familiarize students with the hardware and software architecture of the Cray XC Series computer systems, Cray’s performance analysis tools, and optimization of user codes, through lectures and practical laboratory exercises. The course is offered in-person or remotely.
For more information on the course, and to register, please see https://www.nersc.gov/users/training/events/cori-phase-1-training-programming-and-optimization/ . Note that registration for in-person attendance closes on June 3.
This is the first announcement of the Summer School on Machine Learning for High Energy Physics 2016, to be held at Lund, Sweden, June 20-26 2016 as a satellite event of LHC Physics Conference.
Early registration deadline is 30 April, 2016. Continue reading Machine Learning for High Energy Physics 2016: Lund, Sweden, June 20-26
The National Energy Research Scientific Computing Center (NERSC) has recently announced the selection of a number of science projects for its NERSC Exascale Scientific Applications Program (NESAP), a collaborative effort in which NERSC will partner with code teams to prepare for the NERSC-8 Cori system. Continue reading NESAP Projects