BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211208T164406Z
LOCATION:Level 2 - Exhibit Hall
DTSTART;TZID=America/Los_Angeles:20211206T170000
DTEND;TZID=America/Los_Angeles:20211206T180000
UID:dac_DAC 2021_sess249_des231@linklings.com
SUMMARY:Utilizing the Cloud to Increase Library Characterization Throughpu
 t and Reduce Schedule Bottlenecks
DESCRIPTION:Designer, IP and Embedded Systems Track Poster Networking Rece
 ption\n\nUtilizing the Cloud to Increase Library Characterization Throughp
 ut and Reduce Schedule Bottlenecks\n\nTan, Digraskar\n\nToday, SPICE-based
  characterization of Liberty (.lib) models for modern standard cell librar
 ies involves running hundreds of millions of SPICE simulations, with the a
 mount of simulations needed increasing exponentially over the past few yea
 rs, due to added quantity and complexity of data in .lib models.\n\nDue to
  the high amount of computation required, library characterization teams r
 outinely utilize hundreds to thousands of CPUs during each characterizatio
 n run, and require days or weeks of schedule time to complete each charact
 erization iteration for a given set of .libs. This presents a significant 
 challenge for library teams and IT infrastructure teams trying to meet des
 ign schedules while balancing computational hardware costs.\n\nIn this pap
 er, we present the steps taken and the choices made in setting up a cloud 
 environment for characterization workloads, and quantify scaling efficienc
 y results from a validated testcase. Overall, we find that with the correc
 t configuration, cloud is a valid method providing “burst” com
 pute power to accelerate high priority workloads, while being able to off-
 load on-premises tasks when needed.\n\nVirtual Program: Presented In-Perso
 n
END:VEVENT
END:VCALENDAR