BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191206T142453Z
LOCATION:Substanz 1\, 2
DTSTART;TZID=Europe/Stockholm:20180626T083000
DTEND;TZID=Europe/Stockholm:20180626T100000
UID:isc_hpc_ISC High Performance 2018_sess113_post106@linklings.com
SUMMARY:(RP02) Early Evaluation of a New Vector Processor SX-Aurora TSUBAS
 A
DESCRIPTION:Research Poster\n\n(RP02) Early Evaluation of a New Vector Pro
 cessor SX-Aurora TSUBASA\n\nKomatsu, Momose, Isobe, Sato, Musa...\n\nA bra
 nd-new vector supercomputer, SX-Aurora TSUBASA, has been launched. It has 
 a newly developed Vector Engine(VE) processor. VE is developed to achieve 
 a high sustained performance by powerful vector processing and a high memo
 ry bandwidth.\n\nVE is equipped with eight vector-cores, each of which exe
 cutes 256 operations per vector instruction. As the peak performance of ea
 ch core is 307.2 GFlop/s at 1.6GHz, the peak double-precision performance 
 of VE reaches 2.45 TFlop/s.\n\nVE and six HBM2 memory modules are integrat
 ed by using a CoWoS (Chip on Wafer on Substrate) technology to provide bot
 h high memory bandwidth and large memory capacity, which is the world-firs
 t integration. The memory provides a 1.22 TB/s memory bandwidth and a 48GB
  memory capacity. This memory capability enables a high computational effi
 ciency especially for memory-intensive applications.\n\nFurthermore, SX-Au
 rora TSUBASA employs a new execution model. VE is connected to an X86 serv
 er called Vector Host(VH) via PCIe. Different from the conventional accele
 rator execution model, a whole application is executed on VE, and system c
 alls are automatically offloaded into VH. This new model improves usabilit
 y and avoids special programming for SX-Aurora TSUBASA.\n\nThis paper exam
 ines the potential of SX-Aurora TSUBASA through the performance evaluation
 s. SX-Aurora TSUBASA achieves about 987 GB/s stream performance, which is 
 about 4.7 and 11.6 times higher than the previous vector processor SX-ACE 
 and Xeon Gold 6126, respectively. The high sustained memory bandwidth cont
 ributes 2.0 to 9.8 times performance improvements on application kernels. 
 These results indicate the high potential of SX-Aurora TSUBASA for acceler
 ating memory-intensive applications.\n\nPasses: Conference Pass, Computer 
 Architecture, HPC Accelerators, Heterogeneous Systems, Performance Analysi
 s and Optimization\n\nTag: Conference Pass, Computer Architecture, HPC Acc
 elerators, Heterogeneous Systems, Performance Analysis and Optimization
URL:https://2018.isc-program.com/presentation/?id=post106&sess=sess113
END:VEVENT
END:VCALENDAR