BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000603Z
LOCATION:
DTSTART;TZID=America/Denver:20231115T100000
DTEND;TZID=America/Denver:20231115T150000
UID:submissions.supercomputing.org_SC23_sess503_job168@linklings.com
SUMMARY:NESAP for Simulations Postdoctoral Fellow
DESCRIPTION:NESAP for Simulations Postdoctoral Fellow - 95146\nDivision: N
 E-NERSC\n\nThe National Energy Research Scientific Computing Center (NERSC
 , http://www.nersc.gov/about/) at Berkeley Lab seeks highly motivated post
 docs to join the NERSC Exascale Science Application Program (NESAP, https:
 //www.nersc.gov/users/application-performance/nesap/). NESAP postdocs coll
 aborate with scientific teams to enable the solution of deep, meaningful p
 roblems across all program areas funded by the Department of Energy Office
  of Science.\n\nThe Challenge: Enabling simulation of complex physical sys
 tems, advanced data analytics, and machine learning at scale on energy-eff
 icient supercomputers. Successful candidates will join different teams to 
 focus on one of the following frontiers:\n\nNESAP for Simulations (N4S): C
 utting-edge simulation of complex physical phenomena requires increasing a
 mounts of computational resources due to factors such increasing model siz
 es, parameter space searches, and inclusion of additional physics. N4S ena
 bles simulations to make effective use of modern high-performance computin
 g (HPC) platforms by focusing on algorithm and data structure development 
 and implementation on new architectures such as GPUs, exposing additional 
 parallelism and improving scalability.\n\nTo enable new discoveries throug
 h simulation, data analytics, and ML/DL, NERSC will begin deploying “Perlm
 utter,” a Cray supercomputer, in 2020. Perlmutter, a system optimized for 
 science, is a heterogeneous system including future-generation AMD CPUs an
 d next-generation NVIDIA GPUs. It also has a high-speed interconnect and a
 n all-flash file system.\n\nPerlmutter is NERSC’s first production GPU-bas
 ed system. Many codes running at NERSC need to be adapted or optimized to 
 run efficiently on GPUs. At the same time, solutions that put GPU performa
 nce in users’ hands need to be portable ones. NESAP is about employing cut
 ting-edge computer science and advanced performance analysis tools to deve
 lop highly scalable, distributed parallel algorithms to meet this challeng
 e.\n\nIn this exciting role, you will be a part of a multidisciplinary tea
 m composed of computational and domain scientists working together to tran
 sition and optimize codes to the Perlmutter system and produce mission-rel
 evant science that pushes the limits of HPC. You will carry out code trans
 ition efforts in collaboration with a project PI and team members with the
  support of NERSC and vendor staff. Successful candidates are expected to 
 collaborate with each other across NESAP program areas (N4S, N4D, N4L).\n\
 nNESAP has established a track record of enabling its postdocs to pursue c
 areers in data science, HPC, and scientific computing both in industry and
  at national labs. Take a look at what current and former NESAP postdocs a
 re up to here.\n\nWhat You Will Do:\n• Work with NERSC staff and code team
 s to transition and optimize simulation, data analytics, or machine learni
 ng codes for the Perlmutter system in performance-portable ways.\n• Conduc
 t profiling and scaling studies as well as parallelization, memory bandwid
 th, and I/O analyses for these codes; identify and capitalize on NERSC’s c
 ombined HPC/data ecosystems.\n• Working with domain experts, develop, adap
 t, and optimize state-of-the-art ML/DL models to solve scientific problems
  on HPC systems.\n• Disseminate results of research activities through ref
 ereed publications, reports, and conference presentations. Ensure that new
  methods are documented for the broader community, NERSC staff, vendors, a
 nd NERSC users.\n• Participation in postdoctoral career and science enrich
 ment activities within the Berkeley Lab Computing Sciences Area is encoura
 ged.\n• Opportunities to travel to sites at other labs, universities, and 
 to vendor facilities.\n\n
END:VEVENT
END:VCALENDAR