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:20250626T234540Z
LOCATION:B309
DTSTART;TZID=America/New_York:20241117T120000
DTEND;TZID=America/New_York:20241117T120500
UID:submissions.supercomputing.org_SC24_sess736_ws_pdswwip105@linklings.co
 m
SUMMARY:Scalable RPC Layer Towards Millions of IOPS per Server
DESCRIPTION:Hiroki Ohtsuji, Munenori Maeda, Reika Kinoshita, and Masahiro 
 Miwa (Fujitsu Ltd) and Osamu Tatebe (University of Tsukuba, Japan)\n\nThe 
 growing demand for large-scale AI applications brings performance challeng
 es to parallel file systems. The performance of parallel file systems depe
 nds on both hardware components and software architecture. In parallel fil
 e systems, the performance of remote function execution is critical becaus
 e most operations on a parallel file system require network communication 
 and remote I/O. This work-in-progress paper introduces a new RPC layer tha
 t employs an optimized architecture for many-core CPUs and high-speed netw
 ork devices. The RPC layer adopts a scalable task-stealing model that offe
 rs fairness in task execution and leverages the parallel performance of ma
 ny-core CPUs. Our preliminary performance implementation indicates that th
 e RPC layer can process more than four million RPC operations per second o
 n a single server. This paper introduces the design of the RPC layer and s
 everal performance evaluation results.\n\nTag: Data Movement and Memory, I
 /O, Storage, Archive\n\nRegistration Category: Workshop Reg Pass\n\nSessio
 n Chairs: Suren Byna (The Ohio State University, Lawrence Berkeley Nationa
 l Laboratory (LBNL)); Dean Hildebrand (Google LLC); Anthony Kougkas (Illin
 ois Institute of Technology, Argonne National Laboratory (ANL)); Jay Lofst
 ead (Sandia National Laboratories, University of New Mexico); and Bing Xie
  (Microsoft Corporation)\n\n
END:VEVENT
END:VCALENDAR