BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074357Z
LOCATION:Rio Room
DTSTART;TZID=Europe/Stockholm:20220629T110000
DTEND;TZID=Europe/Stockholm:20220629T130000
UID:submissions.pasc-conference.org_PASC22_sess163@linklings.com
SUMMARY:MS5E - Challenges in Fine-Grained Task Scheduling on Exascale Hete
 rogeneous Architectures: A Molecular Dynamics Perspective
DESCRIPTION:Minisymposium\n\nScientific and technical computing has been f
 acing challenges posed by hardware trends characterized by increasing hard
 ware parallelism, cost of data movement, specialization and heterogeneity.
  To fully utilize the potential of upcoming hardware, efficient task paral
 lelization frameworks are required. Successful frameworks need to be capab
 le of efficiently expressing fine-grained concurrency, scheduling microsec
 ond-granularity tasks and data movement, and tightly integrate with commun
 ication libraries to allow critical-path optimizations in a fully asynchro
 nous schedule. Most of the current tasking frameworks have either struggle
 d to provide low overhead, or have been mainly targeting accelerators, and
  integration with communication standards like MPI is still an active rese
 arch area. Molecular dynamics (MD) has been facing these challenges for a 
 decade due to the typically fixed problem size motivating the strong scali
 ng needs, and very high, sub-millisecond, iteration rates. Hence, the expe
 rience in this field can be beneficial to the broader community likely to 
 face similar challenges on future exascale architectures. In this session 
 experts in runtimes and developers of flagship MD codes GROMACS and NAMD w
 ill to provide a cross-disciplinary view of state-of-the-art in tasking fr
 ameworks, APIs, as well as algorithmic and implementation advances made in
  tackling fine-grained task scheduling challenges while strong-scaling MD 
 on heterogeneous architectures.\n\nDesigning MD Simulations for Modern Het
 erogenous Parallel Architectures\n\nGray, Páll\n\nCutting-edge molecular d
 ynamics (MD) simulations strive for quickest possible evolution of fixed-s
 ized biological systems, requiring strong-scaling across all available het
 erogenous parallel resources in modern accelerated systems. We will showca
 se such achievements for the popular GROMACS MD packag...\n\n-------------
 --------\nPanel Session on Fine-Grained Task Scheduling on Exascale Hetero
 geneous Architectures\n\nHess\n\nThis session will consist of a panel dsic
 ussion of ongoing developments and future directions in fine-grained task 
 scheduling on current and future heterogeneous HPC architectures.\n\n-----
 ----------------\nLeveraging Heterogeneous Architectures for Molecular Dyn
 amics at Exascale\n\nHardy\n\nHeterogeneous architectures at exascale pres
 ent significant challenge to applications like molecular dynamics that are
  latency bound while needing to achieve strong scaling on fixed size probl
 ems. The NAMD parallel molecular dynamics application was from inception d
 esigned for massively distributed ...\n\n---------------------\nTools and 
 Mechanisms to Deal with Fine-Grained Computations\n\nGarcia-Gasulla\n\nWhi
 ch tools and mechanisms can HPC application developers use to deal with fi
 ne-grained computation? In the current context of HPC systems growing in s
 ize and computational power, HPC applications must deal with the challenge
  of using them efficiently. The computing units are becoming more powerful
  ...\n\n\nDomain: Chemistry and Materials, Computer Science and Applied Ma
 thematics, Life Sciences, Physics
END:VEVENT
END:VCALENDAR