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:20211207T054807Z
LOCATION:227-228
DTSTART;TZID=America/Chicago:20211117T140000
DTEND;TZID=America/Chicago:20211117T143000
UID:submissions.supercomputing.org_SC21_sess146_pap302@linklings.com
SUMMARY:Accelerating Applications using Edge Tensor Processing Units
DESCRIPTION:Paper\n\nAccelerating Applications using Edge Tensor Processin
 g Units\n\nHsu, Tseng\n\nNeural network (NN) accelerators have been integr
 ated into a wide range of computer systems. NN accelerators provide native
  hardware support for operations on multidimensional tensor data. Therefor
 e, NN accelerators are theoretically tensor processors that can improve sy
 stem performance for any problem using tensors as inputs and outputs.\n\nT
 his paper introduces General-Purpose Computing on Tensor Processing Units 
 (GPTPU), an open-source, open-architecture framework that allows the devel
 oper and research communities to discover opportunities that NN accelerato
 rs enable for applications. GPTPU includes a powerful programming interfac
 e with efficient runtime system-level support; similar to that of CUDA and
  OpenCL in GPGPU computing; to bridge the gap between application demands 
 and mismatched hardware/software interfaces.\n\nWe built GPTPU machine usi
 ng Edge Tensor Processing Units (Edge TPUs). By leveraging the underlying 
 Edge TPUs to perform main compute kernels, our results reveal that GPTPU a
 chieves a 2.06× speedup over high-end CPUs and reduces energy consumption 
 by 90%.\n\nTag: Reproducibility Badge, Algorithms, Extreme Scale Computing
 , Heterogeneous Systems\n\nRegistration Category: Tech Program Reg Pass\n\
 nReproducibility Badges: Artifact Available, Artifact Functional, Results 
 Reproduced
END:VEVENT
END:VCALENDAR