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:20241120T082409Z
LOCATION:HG F 3
DTSTART;TZID=Europe/Stockholm:20240605T113000
DTEND;TZID=Europe/Stockholm:20240605T133000
UID:submissions.pasc-conference.org_PASC24_sess154@linklings.com
SUMMARY:MS6B - Motif-Based Automated Performance Engineering for HPC
DESCRIPTION:Minisymposium\n\nWe will describe here domain-specific librari
 es (DSLs) that express mathematical/programming motifs (data objects and o
 perations on those data objects), along with software back-ends that trans
 late the library calls into high-performance code. By the use of a motif-a
 ware software stack, the scientific application code written is much small
 er than fully optimized code, with the applications-level code remaining u
 nchanged in moving between platforms, thus leading to a less expensive dev
 elopment process. The four talks being given cover multiple motifs, and di
 fferent approaches to supporting motif-based DSLs. (1) George Bisbas (ICL)
  will talk about an approach to structured-grid DSLs based on lowering the
  abstractions written in Python to the LLVM Multi-Level Intermediate Repre
 sentation (MLIR). (2) Het Mankad (CMU / ORNL) will talk about Proto / Prot
 oX, a DSL for the structured-grid motif that targets CPUs and GPUs, based 
 on the Spiral toolchain. (3) Sanil Rao (CMU) will talk about FFTX for supp
 orting FFTs on CPU and GPU systems, based on the Spiral toolchain. (4) Sam
  Reeve (ORNL) will talk about Cabana, a DSL for supporting grid free parti
 cle methods and hybrid particle / mesh methods on GPUs, based on the use o
 f the Kokkos run-time libraries for GPU parallelism.\n\nProtoX : A Code Ge
 neration Framework for Stencil Operations\n\nProtoX is a code generation f
 ramework for stencil and pointwise operations - the key components in nume
 rically approximating the solution to various partial differential equatio
 ns (PDEs). The frontend for ProtoX uses Proto - a C++ based domain specifi
 c library that provides a high level of abstracti...\n\n\nHet Yagnesh Mank
 ad (Oak Ridge National Laboratory), Sanil Rao (Carnegie Mellon University)
 , Phillip Colella and Brain Van Straalen (Lawrence Berkeley National Labor
 atory), and Franz Franchetti (Carnegie Mellon University)\n---------------
 ------\nFFTX, SpectralPack and Beyond\n\nWe present the design of the API 
 and runtime environment of FFTX as well as future project directions. FFTX
  is developed as part of the DOE ExaScale effort by LBL, Carnegie Mellon U
 niversity, and SpiralGen, Inc. We aim at translating the LAPACK/BLAS appro
 ach from the numerical linear algebra world t...\n\n\nSanil Rao (Carnegie 
 Mellon University)\n---------------------\nScalable and Performance Portab
 le Particle and Structured-Mesh Simulation with Cabana\n\nWe present Caban
 a, a performance portable library for building scientific applications, in
 cluding mesh-free techniques from atomistic (molecular dynamics) to cosmol
 ogy (N-body), hybrid particle-mesh (e.g. particle-in-cell), and structured
  grid simulation. Cabana was created through the U.S. Departme...\n\n\nSam
  Reeve, Lance Bullerwell, Kwitae Chong, David Joy, John Coleman, Pablo Sel
 eson, Steve DeWitt, Matt Rolchigo, Jamie Stump, Wenjun Ge, Tim Younkin, an
 d Stuart Slattery (Oak Ridge National Laboratory)\n---------------------\n
 A Shared Compilation Stack for HPC Stencil DSLs\n\nDomain Specific Languag
 es can massively improve computational science\nproductivity and also prov
 ide high performance. High-level DSLs capture\nabstractions that the DSL c
 ompiler can exploit\nto target current- and next-generation supercomputers
 .\nConsequently, there have been many DSL projects, notably...\n\n\nGeorge
  Bisbas (Imperial College London); Anton Lydike, Emilien Bauer, Nick Brown
 , and Mathieu Fehr (University of Edinburgh); Paul H.J. Kelly (Imperial Co
 llege London); and Tobias Grosser (University of Cambridge)\n\nDomain: Com
 putational Methods and Applied Mathematics\n\nSession Chair: Phillip Colel
 la (Lawrence Berkeley National Laboratory, University of California)
END:VEVENT
END:VCALENDAR
