BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20200129T163556Z
LOCATION:301-302-303
DTSTART;TZID=America/Denver:20191122T091000
DTEND;TZID=America/Denver:20191122T093500
UID:submissions.supercomputing.org_SC19_sess131_ws_ftxs101@linklings.com
SUMMARY:Enforcing Crash Consistency of Scientific Applications in Non-Vola
 tile Main Memory Systems
DESCRIPTION:Workshop\n\nEnforcing Crash Consistency of Scientific Applicat
 ions in Non-Volatile Main Memory Systems\n\nCoy, Zhang\n\nTo fully leverag
 e the emerging non-volatile main memory (NVMM) for high-performance comput
 ing, programmers need efficient data structures that are aware of NVMM mem
 ory models and provide crash consistency. Manual creation of NVMM-aware pe
 rsistent data structures requires a deep understanding of how and when to 
 create persistent snapshots of memory objects corresponding to the data st
 ructures and substantial code modification, which makes it very difficult 
 to use in its manual form even for experienced programmers. To simplify th
 e process, we design a compiler-assistant technique, NVPath. With the aid 
 of compilers, it automatically generates NVMM-aware persistent data struct
 ures which provide the same level of guarantee of crash consistency compar
 ed to the baseline code. Compiler-assistant code annotation and transforma
 tion is general and can be applied to applications using various data stru
 ctures. Furthermore, it is a gray-box technique which requires minimum use
 rs’ input. Finally, it keeps the baseline code structure for good readabil
 ity and maintenance. Our experimental results with real-world scientific a
 pplications (e.g., matrix multiplication, LU decomposition, adaptive-mesh 
 refinement, and page ranking) show that the performance of annotated progr
 ams is commensurate with the version using the manual code transformation 
 on the Titan supercomputer.\n\nTag: Workshop Reg Pass, Extreme Scale Compu
 ting, Fault Tolerance, Reliability, Resiliency\n\nRegistration Category: W
 orkshop Reg Pass, Extreme Scale Computing, Fault Tolerance, Reliability, R
 esiliency
URL:https://sc19.supercomputing.org/presentation/?id=ws_ftxs101&sess=sess1
 31
END:VEVENT
END:VCALENDAR