BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Denver
BEGIN:DAYLIGHT
DTSTART:20240310T030000
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:MDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241103T010000
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:MST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240518T213509Z
UID:07FB56CF-4B36-4AA3-B64A-224AE8CDBAB3
DTSTART;TZID=America/Denver:20240501T180000
DTEND;TZID=America/Denver:20240501T193000
DESCRIPTION:Abstract:\n\nCyber-physical systems (CPS)\, such as robots and 
 self-driving cars\, demand rigorous scheduling to prevent failure\, where 
 every millisecond of processing time can be critical. These systems often 
 rely on heterogeneous computing environments\, which include CPUs\, GPUs\,
  and specialized accelerators\, to meet their computational needs efficien
 tly. However\, leveraging these diverse processing units to fulfill strict
  physical constraints remains a significant challenge\, as existing schedu
 ling solutions often fall short of addressing the complexities involved in
  a comprehensive manner.\n\nThis talk by Dr. Mehmet E. Belviranli delves i
 nto the intricate world of creating efficient compute schedules for CPS th
 at not only cater to their diverse computational hardware but also adhere 
 to real-world constraints critical for system safety. We begin by examinin
 g the role of neural network (NN) inference in CPS\, exploring strategies 
 to balance energy consumption\, latency and throughput by distributing the
  layers of NN across different accelerators. We then introduce a novel\, e
 nd-to-end framework that integrates physical constraints\, heterogeneous c
 omputational resources\, and latency considerations into a cohesive mixed-
 integer linear problem\, demonstrating through case studies how this appro
 ach yields optimal scheduling solutions under varied conditions.\n\nThroug
 h this exploration\, we aim to shed light on the untapped potential of het
 erogeneous computing in enhancing the reliability and performance of CPS. 
 We will also outline future directions in developing a more robust ecosyst
 em for these complex computing environments\, highlighting our contributio
 n to this evolving field.\n\nSpeaker(s): Dr. Belviranli\n\nVirtual: https:
 //events.vtools.ieee.org/m/417666
LOCATION:Virtual: https://events.vtools.ieee.org/m/417666
ORGANIZER:smehalingam@binghamton.edu
SEQUENCE:29
SUMMARY:From Constraints to Capabilities: Achieving Optimal Compute Schedul
 es in Heterogeneous Cyber-Physical Systems
URL;VALUE=URI:https://events.vtools.ieee.org/m/417666
X-ALT-DESC:Description: &lt;br /&gt;&lt;p&gt;&lt;strong&gt;Abstract:&lt;/strong&gt;&lt;/p&gt;\n&lt;p&gt;Cyber-p
 hysical systems (CPS)\, such as robots and self-driving cars\, demand rigo
 rous scheduling to prevent failure\, where every millisecond of processing
  time can be critical. These systems often rely on heterogeneous computing
  environments\, which include CPUs\, GPUs\, and specialized accelerators\,
  to meet their computational needs efficiently. However\, leveraging these
  diverse processing units to fulfill strict physical constraints remains a
  significant challenge\, as existing scheduling solutions often fall short
  of addressing the complexities involved in a comprehensive manner.&lt;/p&gt;\n&lt;
 p&gt;This talk by Dr. Mehmet E. Belviranli delves into the intricate world of
  creating efficient compute schedules for CPS that not only cater to their
  diverse computational hardware but also adhere to real-world constraints 
 critical for system safety. We begin by examining the role of neural netwo
 rk (NN) inference in CPS\, exploring strategies to balance energy consumpt
 ion\, latency and throughput by distributing the layers of NN across diffe
 rent accelerators. We then introduce a novel\, end-to-end framework that i
 ntegrates physical constraints\, heterogeneous computational resources\, a
 nd latency considerations into a cohesive mixed-integer linear problem\, d
 emonstrating through case studies how this approach yields optimal schedul
 ing solutions under varied conditions.&lt;/p&gt;\n&lt;p&gt;Through this exploration\, 
 we aim to shed light on the untapped potential of heterogeneous computing 
 in enhancing the reliability and performance of CPS. We will also outline 
 future directions in developing a more robust ecosystem for these complex 
 computing environments\, highlighting our contribution to this evolving fi
 eld.&lt;/p&gt;\n&lt;p&gt;&amp;nbsp\;&lt;/p&gt;\n&lt;p&gt;&amp;nbsp\;&lt;/p&gt;\n&lt;p&gt;&amp;nbsp\;&lt;/p&gt;
END:VEVENT
END:VCALENDAR