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:20211207T054814Z
LOCATION:230-231-232
DTSTART;TZID=America/Chicago:20211118T163000
DTEND;TZID=America/Chicago:20211118T170000
UID:submissions.supercomputing.org_SC21_sess167_pap362@linklings.com
SUMMARY:Whale: Efficient One-to-Many Data Partitioning in RDMA-Assisted Di
 stributed Stream Processing Systems
DESCRIPTION:Paper\n\nWhale: Efficient One-to-Many Data Partitioning in RDM
 A-Assisted Distributed Stream Processing Systems\n\nTan, Chen, Wang, Jin\n
 \nThe one-to-many data partitioning strategy in a distributed stream proce
 ssing system (DSPS) plays an important role in various applications, where
  the upstream processing instance sends a tuple to a potentially large num
 ber of downstream processing instances. Therefore, a DSPS actually sends a
  same data item to a machine multiple times, raising significant unnecessa
 ry costs for serialization and communications, leading to performance bott
 leneck. \n\nTo address the problem, we design and implement Whale, an effi
 cient RDMA-assisted distributed stream processing system. Whale proposes a
  novel RDMA-assisted stream multicast scheme with a self-adjusting non-blo
 cking tree structure to alleviate the CPU workloads of an upstream instanc
 e during data partitioning. We re-design the DSPS communication mechanism 
 by replacing the instance-oriented communication with a worker-oriented co
 mmunication scheme, which saves significant costs for redundant serializat
 ion and communications. Experimental results show that Whale achieves 56.6
 x improvement of system throughput and 97% reduction of processing latency
  compared to existing designs.\n\nTag: Reproducibility Badge, Accelerator-
 based Architectures, Resource Management and Scheduling\n\nRegistration Ca
 tegory: Tech Program Reg Pass\n\nReproducibility Badges: Artifact Availabl
 e
END:VEVENT
END:VCALENDAR