BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250626T234541Z
LOCATION:B309
DTSTART;TZID=America/New_York:20241117T161000
DTEND;TZID=America/New_York:20241117T161500
UID:submissions.supercomputing.org_SC24_sess736_ws_pdswwip109@linklings.co
 m
SUMMARY:Improving SQL Query Execution of Distributed Query Engines on Obje
 ct-Based Computational Storage through Multi-Layered Offloading
DESCRIPTION:Soon Hwang, Junhyeok Park, and Junghyun Ryu (Sogang University
 , South Korea); Jungahn Park, Jeongjin Lee, Jungki Noh, Soonyeal Yang, and
  Woosuk Chung (SK hynix); and Youngjae Kim (Sogang University, South Korea
 )\n\nThis paper presents an approach to optimize SQL query execution in di
 stributed engines using Object-Based Computational Storage (OCS). Modern a
 nalytics platforms like Presto often suffer from excessive data movement b
 etween compute and storage nodes, even when only a small subset of data is
  required. While solutions like S3 SELECT address this by allowing limited
  operations to be offloaded to storage, they are restricted to simple quer
 ies. The OCS system overcomes these limitations by enabling offloading of 
 more complex, platform-independent query plans via Substrait. This work in
 troduces a multi-layered offloading strategy, where query plans are decomp
 osed and distributed between the OCS Front-End (OCSFE) and OCS Array (OCSA
 ), enhancing resource utilization and reducing query latency. Moreover, th
 is paper presents an integration with Presto, which allows seamless query 
 offloading, and a heuristic algorithm that dynamically manages query distr
 ibution across OCS layers to ensure efficient execution and scalability.\n
 \nTag: Data Movement and Memory, I/O, Storage, Archive\n\nRegistration Cat
 egory: Workshop Reg Pass\n\nSession Chairs: Suren Byna (The Ohio State Uni
 versity, Lawrence Berkeley National Laboratory (LBNL)); Dean Hildebrand (G
 oogle LLC); Anthony Kougkas (Illinois Institute of Technology, Argonne Nat
 ional Laboratory (ANL)); Jay Lofstead (Sandia National Laboratories, Unive
 rsity of New Mexico); and Bing Xie (Microsoft Corporation)\n\n
END:VEVENT
END:VCALENDAR