BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055408Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T090000
DTEND;TZID=America/Chicago:20211114T173000
UID:submissions.supercomputing.org_SC21_sess426@linklings.com
SUMMARY:MCHPC’21: Workshop on Memory Centric High Performance Computing
DESCRIPTION:Workshop\n\nMCHPC'21 Invited Talk: The Path to Memory-Centric 
 Computing in ISO C++\n\nTrott\n\nTo make memory-centric compute devices im
 pactful and sustainable for a wide range of customers in HPC and beyond, w
 e need to find ways of leveraging their capabilities in language standards
 . One of the prime targets for such efforts is the ISO C++ standard, which
  has been the language of choice for ...\n\n---------------------\nMCHPC’2
 1: Workshop on Memory Centric High Performance Computing\n\nYan, Brightwel
 l, Gokhale, Sun\n\nThe growing disparity between CPU speed and memory spee
 d, known as the memory wall problem, has been one of the most critical and
  long-standing challenges in the computing industry. The situation is furt
 her complicated by the recent expansion of the memory hierarchy, which is 
 becoming deeper and mor...\n\n---------------------\nMCHPC'21 Paper Presen
 tation Session I: Advanced Techniques for Using Heterogeneous Memory\n\nBr
 ightwell\n\n---------------------\nMemory Optimizations for Sparse Linear 
 Algebra on GPU Hardware\n\nWalden, Zubair, Stone, Nielsen\n\nAn effort to 
 maximize memory bandwidth utilization\nfor a sparse linear algebra kernel 
 executing on NVIDIA\nTesla V100 and A100 Graphics Processing Units (GPUs) 
 is\ndescribed. The kernel consists of a block-sparse matrix-vector\nproduc
 t and a series of forward/backward triangular solves. The\ncomputation...\
 n\n---------------------\nOpening Remark: MCHPC'21: Workshop on Memory Cen
 tric High Performance Computing\n\nYan, Brightwell, Gokhale, Sun\n\n------
 ---------------\nMCHPC’21 Afternoon Invited Talk:  New Trends for sPIN-Bas
 ed In-Network Computing – from Sparse Reductions to RISC-V Acceleration\n\
 nHoefler\n\nAccelerated in-network computations promise significant optimi
 zations ranging from data-movement reductions to specialization opportunit
 ies in processing elements. We show updates within the sPIN (streaming Pro
 cessing in the Network) network accelerator programming model - the "CUDA 
 for networking"....\n\n---------------------\nUsing Bandwidth Throttling t
 o Quantify Application Sensitivity to Heterogeneous Memory\n\nFoyer, Gogli
 n\n\nMemory management is getting increasingly harder but also primarily i
 mportant. The plurality of processing- and memory-systems require more car
 e to be put into data placement. To test models, designs and heuristics fo
 r data placement, the programmer must be able to access these expensive sy
 stems, or...\n\n---------------------\nMCHPC’21 Morning Break (10-10:30)\n
 \n\n\n---------------------\nPerformance and Energy Improvement of the ECP
  Proxy App SW4lite under Various Workloads\n\nWu, Taylor, Lan\n\nEnergy ef
 ficient execution of scientific applications requires insight into how HPC
  systems affect the performance and energy of the applications. In this pa
 per, we conduct experiments to evaluate the performance of SW4lite under v
 arious workloads with two different memory modes on Cray XC40 Theta a...\n
 \n---------------------\nFreeLunch: Compression-based GPU Memory Managemen
 t for Convolutional Neural Networks\n\nPatel, Liu, Guan\n\nRecently, there
  is a trend to develop deeper and wider Convolutional Neural Networks (CNN
 s) to improve task accuracy. Due to this reason, the GPU memory quickly be
 comes the performance bottleneck since its capacity cannot keep up with th
 e increase of the memory requirement of CNN models. Existing so...\n\n----
 -----------------\nMCHPC'21 Paper Presentation Session II: Application and
  Memory Optimization Techniques\n\nHale\n\n---------------------\nMCHPC’21
  – Morning Invited Talk: The Crucial Role of CXL in Future Memory Centric 
 Computing Systems\n\nBrewer, Gokhale\n\n---------------------\nMCHPC’21 Lu
 nch Break (12:00-2:00)\n\n\n\n---------------------\nClosing Remarks\n\nBr
 ightwell\n\n---------------------\nMCHPC’21:  Afternoon Break (3-3:30)\n\n
 \n\n\nTag: Online Only, Architectures, Memory Systems, Parallel Programmin
 g Languages and Models, System Software and Runtime Systems\n\nRegistratio
 n Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR