Accelerate Your Career at a Special Nov. 16 Recruiting Session
Department of Energy national laboratories are home to many of the world’s fastest supercomputers and linked by the world’s fastest network for science. Computer science research, scientific application development and software creation are on the fast track as part of the Exascale Computing Project. And we need talented people to help do the job.
Learn more about internships, early career positions and other opportunities during a special open recruiting session from 11 a.m. to 1:30 p.m. Thursday, Nov. 16 in the DOE booth (613) at SC17. Recruiters from many of the labs will be on hand to discuss open positions and answer your questions.
Please stop by at your convenience and be ready to shift your career into high gear.
Join Engaging Roundtable Discussions
Pull up a Chair and Chat with Experts: The informal roundtable discussions in the DOE booth provide a great opportunity to rest your feet and engage your mind as you talk about interesting problems and new ways to solve them.
Tuesday, Nov. 14
12:15 p.m.: “Deep Learning for Science”
Join Prabhat from NERSC/Lawrence Berkeley National Laboratory in a discussion about applying deep learning to solve problems in astronomy, cosmology, climate, neuroscience, genomics, high energy physics and nuclear science. He will also discuss his collaboration’s recent success in scaling deep learning to 15PF on the NERSC Cori system.
12:15 p.m.: “LDRD Network Research with SCinet”
Join Sandia National Laboratories network staff in a discussion about three Laboratory Directed Research and Development projects with SCinet, the high-performance network supporting the SC conference series.
4:45 p.m.: “Using the Roofline Model and Intel Advisor”
Join Berkeley Lab’s Sam Williams, developer of the Roofline Model, in an informal discussion. This visually intuitive performance model bounds the performance of various numerical models and operations running on multicore, manycore and accelerator processor architectures and can be used to bound performance improvements on future systems with increased compute, cache or DRAM bandwidths. Intel Advisor now includes both the Cache-Aware Roofline Model (CARM) and a preview of an integrated Roofline model that incorporates multiple levels of the memory hierarchy to help developers pinpoint their bottlenecks in the memory hierarchy.
4:45 p.m.: “TAU Performance System and TAU Commander”
Join Sameer Shende of the University of Oregon in a discussion about the TAU Performance System which delivers robust, integrated, portable, and open technology for performance analysis of parallel applications on large-scale, leadership-class HPC machines available today.
Wednesday, Nov. 15
10 a.m.: “User Stories for Coding@Exascale”
Join Elaine Raybourn from Sandia National Laboratories in an informal discussion aimed at generating user stories that address socio-technical challenges associated with empowering people, processes and productivity. The IDEAS-ECP team is using the agile methodology of user stories to identify SC community needs. We invite you to stop by to share your stories and learn how you can use this method in your own research.
10 a.m. “BigPanDA Project”
Join staff from Brookhaven and Oak Ridge national labs to learn more about BigPanDA Project, which addresses the various aspects of the computing resource needs experiments in High Energy and Nuclear Physics, in particular at the Large Hadron Collider. This will evolve in the future when moving from LHC to High Luminosity LHC in seven years from now, when the already exascale levels of data we are processing could increase by a further order of magnitude.
12:15 p.m.: “NERSC’s Big Data Strategy”
Members of the NERSC Data and Analytics Services team will be available to discuss elements of NERSC’s big data strategy, including the center’s production data stack (analytics, management, transfer, workflows, portals and visualization tools), scaling/optimization efforts through the Big Data Center and NESAP for Data programs and our data science engagement strategy.
4:45 p.m.: “Next-gen I/O with Burst Buffers”
Now that burst buffer technology is moving into the mainstream, Debbie Bard, a big data architect at NERSC, will host a discussion of the different ways scientists are taking advantage of supercomputing burst buffers. She will also talk about NERSC’s experience in administering this emerging technology featuring SSD storage layers.