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:20260522T150117Z
LOCATION:D166
DTSTART;TZID=America/Chicago:20181112T090000
DTEND;TZID=America/Chicago:20181112T173000
UID:submissions.supercomputing.org_SC18_sess173@linklings.com
SUMMARY:ESPM2 2018: Fourth International Workshop on Extreme Scale Program
 ming Models and Middleware
DESCRIPTION:Portable and Reusable Deep Learning Infrastructure with Contai
 ners to Accelerate Cancer Studies\n\nAdvanced programming models, domain s
 pecific languages, and scripting toolkits have the potential to greatly ac
 celerate the adoption of high performance computing.  These complex softwa
 re systems, however, are often difficult to install and maintain, especial
 ly on exotic high-end systems.  We consid...\n\n\nGeorge Zaki (Frederick N
 ational Laboratory for Cancer Research)\n---------------------\nWorkshop M
 orning Break\n---------------------\nWorkshop Lunch (on your own)\n-------
 --------------\nESPM2 2018: Closing Remarks\n\nDhabaleswar Panda and Hari 
 Subramoni (Ohio State University) and Karl Schulz (University of Texas)\n-
 --------------------\nExascale Challenges in Across-Node Parallelism for L
 anguages and Runtimes\n\nMachines with peak performance exceeding one exaf
 lop/s are just around the corner, and promises of sustained exaflop/s mach
 ines abound. Are there significant challenges in runtime frameworks and la
 nguages that need to be met to harness the power of these machines?  We wi
 ll examine this question and ...\n\n\nLaxmikant Kale (University of Illino
 is)\n---------------------\nAnalysis of Explicit vs. Implicit Tasking in O
 penMP Using Kripke\n\nDynamic task-based parallelism has become a widely-a
 ccepted paradigm in the quest for exascale computing. In this work, we del
 iver a non-trivial demonstration of the advantages of explicit over implic
 it tasking in OpenMP 4.5 in terms of both expressiveness and performance. 
 We target the Kripke bench...\n\n\nCharles C. Jin and Muthu Baskaran (Rese
 rvoir Labs Inc)\n---------------------\nWorkshop Afternoon Break\n--------
 -------------\nDesign of Data Management for Multi-SPMD Workflow Programmi
 ng Model\n\nAs both the complexity of algorithms and architecture increase
 , development of scientific software becomes a challenge. In order to expl
 oit future architecture, we consider a Multi-SPMD workflow programing mode
 l. Then, data transfer between tasks during computation highly depends on 
 the architecture...\n\n\nThomas Dufaud (University of Versailles) and Miwa
 ko Tsuji and Mitsuhisa Sato (Riken Center for Computational Science)\n----
 -----------------\nAutomatic Generation of High-Order Finite-Difference Co
 de with Temporal Blocking for Extreme-Scale Many-Core Systems\n\nIn this p
 aper we describe the basic idea, implementation and achieved performance o
 f our DSL for stencil computation, Formura, on systems based on PEZY-SC2 m
 any-core processor. Formura generates, from high-level description of the 
 differential equation and simple description of finite-difference ste...\n
 \n\nHideyuki Tanaka (ExaScaler Inc), Youhei Ishihara (Kyoto University), R
 yo Sakamoto and Takashi Nakamura (PEZY Computing), Yasuyuki Kimura (ExaSca
 ler Inc), Keigo Nitadori and Miyuki Tsubouchi (Riken Center for Computatio
 nal Science), and Jun Makino (Kobe University)\n---------------------\nInt
 egration of CUDA Processing within the C++ Library for Parallelism and Con
 currency (HPX)\n\nExperience shows that on today's high performance system
 s, the utilization of different acceleration cards in conjunction with a h
 igh utilization of all other parts of the system is difficult. Future arch
 itectures, like exascale clusters, are expected to aggravate this issue as
  the number of cores a...\n\n\nPatrick Diehl, Hartmut Kaiser, and Thomas H
 eller (Louisiana State University) and Madhavan Seshadri (Nanyang Technolo
 gical University, Singapore)\n---------------------\nHeterogeneous Systems
  and the Road to Exascale for HPC and AI\n\nDaniel Holmes (University of E
 dinburgh)\n---------------------\nA Unified Runtime for PGAS and Event-Dri
 ven Programming\n\nA well-recognized characteristic of extreme scale syste
 ms is that their computation bandwidths far exceed their communication ban
 dwidths.  PGAS runtimes have proven to be effective in enabling efficient 
 use of communication bandwidth, due to their efficient support for short n
 onblocking one-sided me...\n\n\nSri Raj Paul (Rice University), Kun Chen (
 Georgia Institute of Technology), Akihiro Hayashi and Max Grossman (Rice U
 niversity), and Vivek Sarkar (Georgia Institute of Technology)\n----------
 -----------\nAsynchronous Execution of Python Code on Task Based Runtime S
 ystems\n\nDespite advancements in the areas of parallel and distributed co
 mputing, the complexity of programming on High Performance Computing (HPC)
  resources has deterred many domain experts, especially in the areas of ma
 chine learning and artificial intelligence (AI), from utilizing performanc
 e benefits of ...\n\n\nMohammad Tohid, Bibek Wagle, Shahrzad Shirzad, Patr
 ick Diehl, Adrian Serio, Alireza Kheirkhahan, and Parsa Amini (Louisiana S
 tate University); Katy Williams (University of Arizona); Kevin Huck (Unive
 rsity of Oregon); and Steven Brandt and Hartmut Kaiser (Louisiana State Un
 iversity)\n---------------------\nDistributed Memory Futures for Compile-T
 ime, Deterministic-by-Default Concurrency in Distributed C++ Applications\
 n\nFutures are a widely-used abstraction for enabling deferred execution i
 n imperative programs. Deferred execution enqueues tasks rather than expli
 citly blocking and waiting for them to execute. Many task-based programmin
 g models with some form of deferred execution rely on explicit parallelism
  that i...\n\n\nJeremiah Wilke, David Hollman, Cannada Lewis, Aram Markosy
 an, and Nicolas Morales (Sandia National Laboratories)\n------------------
 ---\nIntroduction - ESPM2 2018: Fourth International Workshop on Extreme S
 cale Programming Models and Middleware\n\nNext generation architectures an
 d systems being deployed are characterized by high concurrency, low memory
  per-core, and multiple levels of hierarchy and heterogeneity. These chara
 cteristics bring out new challenges in energy efficiency, fault-tolerance,
  and scalability. It is commonly believed that...\n\n\nHari Subramoni (Ohi
 o State University), Karl Schulz (University of Texas), and Dhabaleswar Pa
 nda (Ohio State University)\n\nTag: Accelerators, Exascale, Parallel Progr
 amming Languages, Libraries, and Models\n\nRegistration Category: Workshop
  Reg Pass\n\nSession Chairs: Dhabaleswar K. (DK) Panda (The Ohio State Uni
 versity), Karl Schulz (Advanced Micro Devices (AMD) Inc), and Hari Subramo
 ni (The Ohio State University)
END:VEVENT
END:VCALENDAR