BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:US/Pacific
BEGIN:DAYLIGHT
DTSTART:20240310T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241103T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240813T201710Z
UID:052EED7B-A4D6-44F1-9106-654B7FC85E35
DTSTART;TZID=US/Pacific:20240717T183000
DTEND;TZID=US/Pacific:20240717T203000
DESCRIPTION:Event Details\n\nPer a recent [Forbes article](https://www.forb
 es.com/sites/bethkindig/2024/06/20/ai-power-consumption-rapidly-becoming-m
 ission-critical/)\, "The rise of generative AI and surging GPU shipments i
 s causing data centers to scale from tens of thousands to 100\,000-plus ac
 celerators\, shifting the emphasis to power as a mission-critical problem 
 to solve." making this talk particularly relevant.\n\nThe US Department of
  Energy projects that by the year 2055\, given current projected growth\, 
 electrical demand for devices including the internet will exceed the capac
 ity of planet Earth to provide that power. The electrical demands on data 
 processing centers in a world of AI and cryptocurrency on top of standard 
 workloads lead the industry down an unsustainable path. Hyperscale systems
  are feverishly adding processing\, memory\, storage\, connectivity fabric
 \, and communications at an alarming rate. Software engineers need to be p
 art of the total cost of ownership calculation to address this energy cris
 is.\nThis talk details some key aspects of the system architectures that a
 re being deployed in data centers\, with focus on data movement between re
 sources. Every time data is moved there is inherent inefficiency\; we will
  calculate the wasted data movement in key parts of the system architectur
 e. With this analysis\, it is clear that the expansion of the data center 
 architecture has exacerbated the variations in access latency time to each
  resource as well.\n\nSoftware developers can leverage knowledge of the en
 ergy flow through the system to improve efficiency. Gains can be made on e
 very level from the choice of machine level data types – to choice of hi
 gh level languages – to various programmer interfaces between the applic
 ation and the hardware resources. The variations in media access latency a
 re only getting wider and more complex\, so software developers are tasked
  with comprehending not only what data to process but where to place it in
  the system hierarchy.\n\nThe call to action will summarize some of the sp
 ecific ways careful software engineering can help us avert the looming pow
 er crisis.\n\nAbout the Speaker Bill Gervasi.\n\nBill Gervasi is Principal
  Systems Architect for Wolley Inc\, designing advanced memory and storage 
 solutions for data\n\nNOTE: This is an IN PERSON meeting. To receive a lin
 k to the event's recording\, complete the survey referenced in the header 
 or footer.\n\nCo-sponsored by: OC ACM [Actual Host] and Knobbe Martens\, a
 n Intellectual Property & Technology law firm [Physical Host Donor]\n\nAge
 nda: \n6:30 PM Networking at physical meeting location\n7:00 PM Announceme
 nts and Presentation with Q&A\n8:00 PM Follow-up quesitons for presenter a
 nd networking\n8:30 PM Meeting Adjourned\n\n2040 Main St \, Ste 14\, Irvin
 e\, California\, United States
LOCATION:2040 Main St \, Ste 14\, Irvine\, California\, United States
ORGANIZER:AWBrown@USC.edu
SEQUENCE:44
SUMMARY:"Systems and Software for Averting an Energy Crisis" IEEE OCCS & OC
  ACM Mtg
URL;VALUE=URI:https://events.vtools.ieee.org/m/426759
X-ALT-DESC:Description: <br /><p><span style="text-decoration: underline\;"
 ><strong>Event</strong> <strong>Details</strong></span><strong> &nbsp\;</s
 trong></p>\n<p>Per a recent <a href="https://www.forbes.com/sites/bethkind
 ig/2024/06/20/ai-power-consumption-rapidly-becoming-mission-critical/">For
 bes article</a>\, "The rise of generative AI and surging GPU shipments is 
 causing data centers to scale from tens of thousands to 100\,000-plus acce
 lerators\, shifting the emphasis to power as a mission-critical problem to
  solve." making this talk particularly relevant.</p>\n<p>The US Department
  of Energy projects that by the year 2055\, given current projected growth
 \, electrical demand for devices including the internet will exceed the ca
 pacity of planet Earth to provide that power. The electrical demands on da
 ta processing centers in a world of AI and cryptocurrency on top of standa
 rd workloads lead the industry down an unsustainable path. Hyperscale syst
 ems are feverishly adding processing\, memory\, storage\, connectivity fab
 ric\, and communications at an alarming rate. Software engineers need to b
 e part of the total cost of ownership calculation to address this energy c
 risis.<br>This talk details some key aspects of the system architectures t
 hat are being deployed in data centers\, with focus on data movement betwe
 en resources. Every time data is moved there is inherent inefficiency\; we
  will calculate the wasted data movement in key parts of the system archit
 ecture. With this analysis\, it is clear that the expansion of the data ce
 nter architecture has exacerbated the variations in access latency time to
  each resource as well.</p>\n<p>Software developers can leverage knowledge
  of the energy flow through the system to improve efficiency. Gains can be
  made on every level from the choice of machine level data types &ndash\; 
 to choice of high level languages &ndash\; to various programmer interface
 s between the application and the hardware resources. The variations in me
 dia access latency are only getting wider and more complex\, so software d
 evelopers are tasked with comprehending not only what data to process but 
 where to place it in the system hierarchy.</p>\n<p>The call to action will
  summarize some of the specific ways careful software engineering can help
  us avert the looming power crisis.</p>\n<p><strong><span style="text-deco
 ration: underline\;">About the Speaker</span> Bill Gervasi.&nbsp\;</strong
 ></p>\n<p>Bill Gervasi is Principal Systems Architect for Wolley Inc\, des
 igning advanced memory and storage solutions for data&nbsp\;</p>\n<p><span
  style="text-decoration: underline\;"><strong>NOTE:</strong></span>&nbsp\;
  This is an IN PERSON meeting.&nbsp\; <strong>To receive a link to the eve
 nt's recording\, complete the survey referenced in the header or footer.</
 strong></p><br /><br />Agenda: <br /><p><span class="aBn" tabindex="0" dat
 a-term="goog_557201553"><span class="aQJ">6:30 PM Networking at physical m
 eeting location<br></span></span><span class="aBn" tabindex="0" data-term=
 "goog_557201553"><span class="aQJ">7:00 PM</span></span> Announcements and
  Presentation with Q&amp\;A<br>8:00 PM Follow-up quesitons for presenter a
 nd networking<br>8:30 PM Meeting Adjourned</p>
END:VEVENT
END:VCALENDAR