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:20211207T055409Z
LOCATION:227-228
DTSTART;TZID=America/Chicago:20211116T153000
DTEND;TZID=America/Chicago:20211116T170000
UID:submissions.supercomputing.org_SC21_sess153@linklings.com
SUMMARY:Networks
DESCRIPTION:Paper\n\nHatRPC: Hint-Accelerated Thrift RPC over RDMA\n\nLi, 
 Shi, Lu\n\nIn this paper, we propose a novel hint-accelerated RPC framewor
 k based on Apache Thrift over Remote Direct Memory Access (RDMA) protocols
 , called HatRPC. HatRPC adopts a hierarchical hint design towards optimizi
 ng heterogeneous RPC services and functions. The proposed hint design is c
 omposed of serv...\n\n---------------------\nSEEC: Stochastic Escape Expre
 ss Channel\n\nParasar, Enright Jerger, Gratz, San Miguel, Krishna\n\nAlloc
 ating a free buffer before moving to the next router is a fundamental tene
 t for packet movement in NoCs. There are two challenges, however. First, i
 f there is a cyclic dependency among the buffers, it can lead to deadlock.
  Second, if buffers are full of packets going towards congested regions, .
 ..\n\n---------------------\nFlare: Flexible In-Network Allreduce\n\nDe Se
 nsi, Di Girolamo, Ashkboos, Li, Hoefler\n\nThe allreduce operation is one 
 of the most commonly used communication routines in distributed applicatio
 ns. To improve its bandwidth and to reduce network traffic, this operation
  can be accelerated by offloading it to network switches, which aggregate 
 the data received from the hosts and send back ...\n\n\nTag: Networks\n\nR
 egistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR