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:20181221T160911Z
LOCATION:D165
DTSTART;TZID=America/Chicago:20181112T085900
DTEND;TZID=America/Chicago:20181112T173000
UID:submissions.supercomputing.org_SC18_sess161@linklings.com
SUMMARY:The 9th International Workshop on Performance Modeling, Benchmarki
 ng, and Simulation of High-Performance Computer Systems (PMBS18)
DESCRIPTION:Workshop\nBenchmarks, Parallel Programming Languages, Librarie
 s, and Models, Performance, Simulation, Workshop Reg Pass\n\nWorkshop Morn
 ing Break\n\n\n\n---------------------\nWorkshop Afternoon Break\n\n\n\n--
 -------------------\nIntroduction - The 9th International Workshop on Perf
 ormance Modeling, Benchmarking, and Simulation of High-Performance Compute
 r Systems (PMBS18)\n\nWright, Jarvis, Hammond\n\nThe PMBS18 workshop is co
 ncerned with the comparison of high-performance computing systems through 
 performance modeling, benchmarking or through the use of tools such as sim
 ulators. We are particularly interested in research which reports the abil
 ity to measure and make tradeoffs in software/hardwar...\n\n--------------
 -------\nA Metric for Evaluating Supercomputer Performance in the Era of E
 xtreme Heterogeneity\n\nAustin, Daley, Doerfler, Deslippe, Cook...\n\nWhen
  acquiring a supercomputer, it is desirable to specify its performance usi
 ng a single number. For many procurements, this is usually stated as a per
 formance increase over a current generation platform, for example machine 
 A provides 10 times greater performance than machine B. The determination 
 ...\n\n---------------------\nApproximating a Multi-Grid Solver\n\nLe Fèvr
 e, Bautista-Gomez, Unsal, Casas\n\nMulti-grid methods are numerical algori
 thms used in parallel and distributed processing. The main idea of multi-g
 rid solvers is to speed up the convergence of an iterative method by reduc
 ing the problem to a coarser grid a number of times. Multi-grid methods ar
 e widely exploited in many application ...\n\n---------------------\nBench
 marking Machine Learning Methods for Performance Modeling of Scientific Ap
 plications\n\nMalakar, Balaprakash, Vishwanat, Morozov, Kumaran\n\nPerform
 ance modeling is an important and active area of research in high-performa
 nce computing (HPC). It helps in better job scheduling and also improves o
 verall performance of coupled applications. Sufficiently rich analytical m
 odels are challenging to develop, however, because of interactions betw...
 \n\n---------------------\nEvaluating SLURM Simulator with Real-Machine SL
 URM and Vice Versa\n\nJokanovic, D'Amico, Corbalan\n\nHaving a precise and
  a fast job scheduler model that resembles the real-machine job scheduling
  software behavior is extremely important in the field of job scheduling. 
 The idea behind SLURM simulator is preserving the original code of the cor
 e SLURM functions while allowing for all the advantages of...\n\n---------
 ------------\nIs Data Placement Optimization Still Relevant on Newer GPUs?
 \n\nBari, Stoltzfus, Lin, Liao, Emani...\n\nModern supercomputers often us
 e Graphic Processing Units (or GPUs) to meet the evergrowing demands for e
 nergy efficient high performance computing. GPUs have a complex memory arc
 hitecture with various types of memories and caches, such as global memory
 , shared memory, constant memory, and texture me...\n\n-------------------
 --\nDeep Learning at Scale on Nvidia V100 Accelerators\n\nXu, Han, Ta\n\nT
 he recent explosion in the popularity of Deep Learning (DL) is due to a co
 mbination of improved algorithms, access to large datasets and increased c
 omputational power. This had led to a plethora of open-source DL framework
 s, each with varying characteristics and capabilities. End users are then 
 lef...\n\n---------------------\nAlgorithm Selection of MPI Collectives Us
 ing Machine Learning Techniques\n\nHunold, Carpen-Amarie\n\nAutotuning is 
 a well established method to improve software performance for a given syst
 em, and it is especially important in High Performance Computing. The goal
  of autotuning is to find the best possible algorithm and its best paramet
 er settings for a given instance. Autotuning can also be applied...\n\n---
 ------------------\nAutomated Instruction Stream Throughput Prediction for
  Intel and AMD Microarchitectures\n\nLaukemann, Hammer, Hofmann, Hager, We
 llein\n\nAn accurate prediction of scheduling and execution of instruction
  streams is a necessary prerequisite for predicting the in-core performanc
 e behavior of throughput-bound loop kernels on out-of-order processor arch
 itectures. Such predictions are an indispensable component of analytical p
 erformance mo...\n\n---------------------\nUnified Cross-Platform Profilin
 g of Parallel C++ Applications\n\nKucher, Fey, Gorlatch\n\nTo address the 
 great variety of available parallel hardware architectures (CPUs, GPUs, et
 c.), high-performance applications increasingly demand cross-platform port
 ability.  While unified programming models like OpenCL or SYCL provide the
  ultimate portability of code, the profiling of applications in...\n\n----
 -----------------\nEvaluating the Impact of Spiking Neural Network Traffic
  on Extreme-Scale Hybrid Systems\n\nWolfe, Plagge, Mubarak, Carothers, Ros
 s\n\nAs we approach the limits of Moore's law, there is increasing interes
 t in non-Von Neuman architectures such as neuromorphic computing to take a
 dvantage of improved compute and low power capabilities. Spiking neural ne
 twork (SNN) applications have so far shown very promising results running 
 on a numb...\n\n---------------------\nExploring and Quantifying How Commu
 nication Behaviors in Proxies Relate to Real Applications\n\nAaziz, Cook, 
 Cook, Vaughan\n\nProxy applications, or proxies, are simple applications m
 eant to exercise systems in a way that mimics real applications (their par
 ents). However, characterizing the relationship between the behavior of pa
 rent and proxy applications is not an easy task. In prior work, we present
 ed a data-driven meth...\n\n---------------------\nWorkshop Lunch (on your
  own)\n\n\n\n---------------------\nImproving MPI Reduction Performance fo
 r Manycore Architectures with OpenMP and Data Compression\n\nShan, William
 s, Johnson\n\nMPI reductions are widely used in many scientific applicatio
 ns and often become the scaling performance bottleneck. When performing re
 ductions on vectors, different algorithms have been developed to balance m
 essaging overhead and bandwidth. However, most implementations have ignore
 d the effect of si...\n\n---------------------\nminiVite: A Graph Analytic
 s Benchmarking Tool for Massively Parallel Systems\n\nGhosh, Halappanavar,
  Tumeo, Kalyanaraman, Gebremedhin\n\nBenchmarking of high performance comp
 uting systems can help provide critical insights for efficient design of c
 omputing systems and software applications. Although a large number of too
 ls for benchmarking exist, there is a lack of representative benchmarks fo
 r the class of irregular computations as ...\n
END:VEVENT
END:VCALENDAR