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:20181221T160725Z
LOCATION:D167/174
DTSTART;TZID=America/Chicago:20181111T111000
DTEND;TZID=America/Chicago:20181111T112500
UID:submissions.supercomputing.org_SC18_sess170_ws_h2rc103@linklings.com
SUMMARY:First Steps in Porting the LFRic Weather and Climate Model to the 
 FPGAs of the EuroExa Architecture
DESCRIPTION:Workshop\nAccelerators, Heterogeneous Systems, NVRAM, Workshop
  Reg Pass\n\nFirst Steps in Porting the LFRic Weather and Climate Model to
  the FPGAs of the EuroExa Architecture\n\nAshworth, Riley, Attwood, Mawer\
 n\nThe EuroExa project proposes a High-Performance Computing (HPC) archite
 cture which is both scalable to exascale performance levels and delivers w
 orld-leading power efficiency. This is achieved through the use of low-pow
 er ARM processors accelerated by closely-coupled FPGA programmable compone
 nts. In order to demonstrate the efficacy of the design, the EuroExa proje
 ct includes application porting work across a rich set of applications. On
 e such application is the new weather and climate model, LFRic (named in h
 onor of Lewis Fry Richardson), which is being developed by the UK Met Offi
 ce and its partners for operational deployment in the middle of the next d
 ecade. \n\nMuch of the run-time of the LFRic model consists of compute int
 ensive operations which are suitable for acceleration using FPGAs. Program
 ming methods for such high-performance numerical workloads are still immat
 ure for FPGAs compared with traditional HPC architectures. The paper descr
 ibes the porting of a matrix-vector kernel using the Xilinx Vivado toolset
 , including High-Level Synthesis (HLS), discusses the benefits of a range 
 of optimizations and reports performance achieved on the Xilinx UltraScale
 + SoC. \n\nPerformance results are reported for the FPGA code and compared
  with single socket OpenMP performance on an Intel Broadwell CPU. We find 
 the performance of the FPGA to be competitive when taking into account pri
 ce and power consumption.
URL:https://sc18.supercomputing.org/presentation/?id=ws_h2rc103&sess=sess1
 70
END:VEVENT
END:VCALENDAR