NERSC Training Events
See also the NERSC Events Calendar.
Scaling to Petascale and Beyond: Performance Analysis and Optimization of Applications Read More »
Speakers: Heidi Poxon, John Levesque, Luiz DeRose, Cray Location: Berkeley Lab, Room 50B-4205. Berkeley, CA Time: 9:00am - 5:00pm PDT, June 14, 2018. Workshop DescriptionThe "Applying a “Whack-a-mole” Method using Cray’s perftools to identify the Moles" presentation in the morning will demonstrate some of the new capabilities available with the Cray performance tools, and will cover a recommended process of using the tools to identify key bottlenecks (moles) in a program, and then… Read More »
Speakers: Timothy Mattson, Intel; Helen He, NERSC Location: Berkeley Lab, Room 50B-1237. Berkeley, CA Time: 9:00am - 5:00pm PDT, May 4, 2018. Tutorial DescriptionMost OpenMP programmers use only about 19 items from the language. We call these 19 items “the OpenMP Common Core”. Programmers may spend their entire OpenMP programming careers safely inside the common core. Well rounded OpenMP programmers, however, should be aware of what lays just beyond the common core. These less… Read More »
NERSC will be hosting an in-person and remote training for all users on Cori KNL nodes. Read More »
NERSC will host a training event for users about Intel compilers, tools, and libraries on Tuesday, March 6, 2018. The main presenters will be from Intel. The training will be held at Building 59, Room 3101 (CRT Building, aka Wang Hall).Note: There is a Cori KNL Hackathon on the following day that we strongly suggest you to register as well, so that you can directly apply the information learned here towards the improving your application performance on Cori KNL.Attendees who have an… Read More »
DescriptionOpenMP is the de facto standard for writing parallel applications for shared memory computers. Over 20 years of the development, with a specification over 350 pages now, OpenMP has grown into an intimidating API viewed by many as for “experts only”.Most OpenMP programmers, however, use around 19 items from the specification. We call these 19 items the “OpenMP Common Core”. By focusing on the common core, we make OpenMP what it was always meant to be; an easy API for parallel… Read More »