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:20260522T150153Z
LOCATION:D163
DTSTART;TZID=America/Chicago:20181112T090000
DTEND;TZID=America/Chicago:20181112T173000
UID:submissions.supercomputing.org_SC18_sess142@linklings.com
SUMMARY:PDSW-DISCS: Joint International Workshop on Parallel Data Storage 
 and Data Intensive Scalable Computing Systems
DESCRIPTION:WIP Session 1\n---------------------\nWorkshop Morning Break\n
 ---------------------\nWorkshop Afternoon Break\n---------------------\nMe
 thodology for the Rapid Development of Scalable HPC Data Services\n\nGrowi
 ng evidence in the scientific computing community indicates that parallel 
 file systems are not sufficient for all HPC storage workloads.  This reali
 zation has motivated extensive research in new storage system designs. The
  question of which design we should turn to implies that there could be a.
 ..\n\n\nMatthieu Dorier, Philip Carns, Kevin Harms, Robert Latham, Robert 
 Ross, Shane Snyder, and Justin Wozniak (Argonne National Laboratory); Samu
 el K. Gutiérrez, Bob Robey, Brad Settlemyer, and Galen Shipman (Los Alamos
  National Laboratory); Jerome Soumagne (HDF Group); and James Kowalkowski,
  Marc Paterno, and Saba Sehrish (Fermi National Accelerator Laboratory)\n-
 --------------------\nPufferbench: Evaluating and Optimizing Malleability 
 of Distributed Storage\n\nMalleability is the property of an application t
 o be dynamically rescaled at run time. It requires the possibility to dyna
 mically add or remove resources to the infrastructure without interruption
 . Yet, many Big Data applications cannot benefit from their inherent malle
 ability, since their colocated...\n\n\nNathanael Cheriere (IRISA, ENS Renn
 es); Matthieu Dorier (Argonne National Laboratory); and Gabriel Antoniu (F
 rench Institute for Research in Computer Science and Automation (INRIA))\n
 ---------------------\nEvaluation of HPC Application I/O on Object Storage
  Systems\n\nPOSIX-based parallel file systems provide strong consistency s
 emantics, which many modern HPC applications do not need and do not want. 
  Object store technologies avoid POSIX consistency and are designed to be 
 extremely scalable, for use in cloud computing and similar commercial envi
 ronments.  In th...\n\n\nJialin Liu, Quincey Koziol, and Gregory F. Butler
  (Lawrence Berkeley National Laboratory); Neil Fortner (HDF Group); Mohama
 d Chaarawi (Intel Corporation); and Houjun Tang, Suren Byna, Glenn K. Lock
 wood, Ravi Cheema, Kristy A. Kallback-Rose, Damian Hazen, and Mr Prabhat (
 Lawrence Berkeley National Laboratory)\n---------------------\nWelcome and
  Introduction\n---------------------\nIntroduction - PDSW-DISCS: Joint Int
 ernational Workshop on Parallel Data Storage and Data Intensive Scalable C
 omputing Systems\n\nWe are pleased to announce that the third Joint Intern
 ational Workshop on Parallel Data Storage and Data Intensive Scalable Comp
 uting Systems (PDSW-DISCS’18) will be hosted at SC18: The International Co
 nference for High Performance Computing, Networking, Storage, and Analysis
 .  The objective of this...\n\n\nKathryn Mohror (Lawrence Livermore Nation
 al Laboratory); Suzanne McIntosh (New York University); Raghu Raja (Amazon
  Web Services); Carlos Maltzahn and Ivo Jimenez (University of California,
  Santa Cruz); Glenn Lockwood (Lawrence Berkeley National Laboratory); and 
 Joan Digney (Carnegie Mellon University)\n---------------------\nWIP Sessi
 on 2\n---------------------\nToward Understanding I/O Behavior in HPC Work
 flows\n\nScientific discovery increasingly depends on complex workflows co
 nsisting of multiple phases and sometimes millions of parallelizable tasks
  or pipelines. These workflows access storage resources for a variety of p
 urposes, including preprocessing, simulation output, and postprocessing st
 eps. Unfortun...\n\n\nJakob Luettgau (German Climate Computing Center, Arg
 onne National Laboratory); Shane Snyder, Philip Carns, and Justin M. Wozni
 ak (Argonne National Laboratory); Julian Kunkel (University of Reading); a
 nd Thomas Ludwig (German Climate Computing Center)\n---------------------\
 nIntegration of Burst Buffer in High-Level Parallel I/O Library for Exasca
 le Computing Era\n\nWhile the computing power of supercomputers continues 
 to improve at an astonishing rate, companion I/O systems are struggling to
  keep up in performance. To mitigate the performance gap, several supercom
 puting systems have been configured to incorporate burst buffers into thei
 r I/O stack; the exact r...\n\n\nKai-Yuan Hou, Reda Al-Bahrani, Esteban Ra
 ngel, and Ankit Agrawal (Northwestern University); Robert Latham and Rober
 t Ross (Argonne National Laboratory); and Alok Choudhary and Wei-keng Liao
  (Northwestern University)\n---------------------\nUsing a Robust Metadata
  Management System to Accelerate Scientific Discovery at Extreme Scales\n\
 nOur previous work, which can be referred to as EMPRESS 1.0, showed that r
 ich metadata management provides a relatively low-overhead approach to fac
 ilitating insight from scale-up scientific applications. However, this sys
 tem did not provide the functionality needed for a viable production syste
 m or ...\n\n\nMargaret Lawson (Sandia National Laboratories, University of
  Illinois) and Jay Lofstead (Sandia National Laboratories)\n--------------
 -------\nWorkshop Lunch (on your own)\n---------------------\nWIP Session 
 3\n---------------------\nCharacterizing Deep-Learning I/O Workloads in Te
 nsorFlow\n\nThe performance of Deep-Learning (DL) computing frameworks rel
 y on the performance of data ingestion and checkpointing. In fact, during 
 the training, a considerable high number of relatively small files are fir
 st loaded and pre-processed on CPUs and then moved to accelerator for comp
 utation. In addi...\n\n\nWei Der Chien, Stefano Markidis, and Chaitanya Pr
 asad Sishtla (KTH Royal Institute of Technology); Luis Santos (Institute S
 uperior Técnico); Pawel Herman (KTH Royal Institute of Technology); Sai Na
 rasimhamurthy (Seagate Systems UK); and Erwin Laure (KTH Royal Institute o
 f Technology)\n---------------------\nKeynote Address\n\nRangan Sukumar (C
 ray Inc)\n---------------------\nUnderstanding SSD Reliability in Large-Sc
 ale Cloud Systems\n\nModern datacenters increasingly use flash-based solid
  state drives (SSDs) for high performance and low energy cost. However, SS
 Ds introduce more complex failure modes compared to traditional hard disks
 . While great efforts have been made to understand the reliability of SSDs
  itself, it remains uncle...\n\n\nErci Xu (Ohio State University), Mai Zhe
 ng (Iowa State University), Feng Qin (Ohio State University), and Yikang X
 u and Jiesheng Wu (Alibaba Inc)\n\nTag: I/O, Storage\n\nRegistration Categ
 ory: Workshop Reg Pass
END:VEVENT
END:VCALENDAR