BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20251102T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20251114T185034Z UID:3CDB8FD5-A577-4E31-A711-F51B8507FA0D DTSTART;TZID=America/New_York:20251113T190000 DTEND;TZID=America/New_York:20251113T203000 DESCRIPTION:Boston Chapter of the IEEE Computer Society\, GBC/ACM and MIT S tudent Chapter of SIAM (Society for Industrial and Applied Mathematics)\n\ n7:00 PM\, Thursday\, 13 November 2025\n\nMIT Room 32-G449 (Kiva) and onli ne via Zoom\n\nEasy Acceleration with Distributed Arrays on the World’s Largest Interactive AI Supercomputer\n\nJeremy Kepner\n\nPlease register i n advance for this seminar even if you plan to attend in person at\n\n<htt ps://acm-org.zoom.us/webinar/register/1617607374551/WN_lYs4lxKfSlGkMVq71ib N-g>\n\nAfter registering\, you will receive a confirmation email containi ng information about joining the webinar.\n\nIndicate on the registration form if you plan to attend in person. This will help us determine whether the room is close to reaching capacity. We plan to serve light refreshment s (probably pizza) before the talk starting at around 6:30 pm. Letting us know you will come in person will help us determine how much pizza to orde r.\n\nWe may make some auxiliary material such as slides and access to the recording available after the seminar to people who have registered.\nAbs tract:\nHigh level programming languages and GPU accelerators are powerful enablers for a wide range of applications. Achieving scalable vertical (w ithin a compute node)\, horizontal (across compute nodes)\, and temporal ( over different generations of hardware) performance while retaining produc tivity requires effective abstractions. Distributed arrays are one such ab straction that enables high level programming to achieve highly scalable p erformance. Distributed arrays achieve this performance by deriving parall elism from data locality\, which naturally leads to high memory bandwidth efficiency. This talk explores distributed array performance on a variety of hardware. Scalable performance is demonstrated within and across CPU co res\, CPU nodes\, and GPU nodes. The interactive AI supercomputing hardwar e used spans decades and allows a direct comparison of hardware improvemen ts over this time range.\n\nBio:\n\nDr. Jeremy Kepner is an MIT Lincoln La boratory Fellow. He founded the Lincoln Laboratory Supercomputing Center a nd pioneered the establishment of the Massachusetts Green High Performance Computing Center. He has developed novel big data and parallel computing software used by thousands of scientists and engineers worldwide. He has l ed several embedded computing efforts\, which earned him a 2011 R&D 100 Aw ard. Kepner has chaired the SIAM Data Mining conference\, the IEEE Big Dat a conference\, and the IEEE High Performance Extreme Computing conference. Kepner is the author of two bestselling books\, Parallel MATLAB for Multi core and Multinode Computers\, and Graph Algorithms in the Language of Lin ear Algebra. His peer-reviewed publications include works on abstract alge bra\, astronomy\, astrophysics\, cloud computing\, cybersecurity\, data mi ning\, databases\, graph algorithms\, health sciences\, plasma physics\, s ignal processing\, and 3D visualization. In 2014\, he received Lincoln Lab oratory's Technical Excellence Award.\n\nKepner holds a BA degree in astro physics from Pomona College and a PhD degree in astrophysics from Princeto n University. He is a fellow of the Society of Industrial Applied Mathemat ics (SIAM) and is a faculty advisor to the MIT SIAM student group.\n\nDire ctions to 32-G449 - MIT Stata Center\, 32 Vassar Street\, Cambridge\, MA: Please use the main entrance to the Stata Center at 32 Vassar Street (the entrance closest to Main street) as those doors will be unlocked. Upon ent ering\, proceed to the elevators which will be on the right after passing a large set of stairs and a MITAC kiosk. Take the elevator to the 4th floo r and turn right\, following the hall to an open area\; 32-G449 will be on the left. [Location of Stata on campus map](https://whereis.mit.edu/?go=3 2)\n\nThis joint meeting of the Boston Chapter of the IEEE Computer Societ y and GBC/ACM will be hybrid (in person and online).\n\nUp-to-date informa tion about this and other talks is available online at https://ewh.ieee.or g/r1/boston/computer/. You can sign up to receive updated status informati on about this talk and informational emails about future talks at https:// mailman.mit.edu/mailman/listinfo/ieee-cs\, our self-administered mailing l ist.\n\nCo-sponsored by: gbc/acm and MIT Student Chapter of SIAM (Society for Industrial and Applied Mathematics)\n\nSpeaker(s): Jeremy Kepner\, \n\ nRoom: MIT Room 32-G449 (Kiva)\, Bldg: Stata Center\, MIT building 32\, 32 Vassar St\, Cambridge\, Massachusetts\, United States\, Virtual: https:// events.vtools.ieee.org/m/508372 LOCATION:Room: MIT Room 32-G449 (Kiva)\, Bldg: Stata Center\, MIT building 32\, 32 Vassar St\, Cambridge\, Massachusetts\, United States\, Virtual: h ttps://events.vtools.ieee.org/m/508372 ORGANIZER:p.mager@computer.org SEQUENCE:33 SUMMARY:Easy Acceleration with Distributed Arrays on the World’s Largest Interactive AI Supercomputer URL;VALUE=URI:https://events.vtools.ieee.org/m/508372 X-ALT-DESC:Description: <br /><blockquote>\n<h2>Boston Chapter of the IEEE Computer Society\, GBC/ACM and MIT Student Chapter of SIAM (Society for In dustrial and Applied Mathematics)</h2>\n<h4>7:00 PM\, Thursday\, 13 Novemb er 2025</h4>\n<p>MIT Room 32-G449 (Kiva) and online via Zoom</p>\n<p>&nbsp \;</p>\n<h4>Easy Acceleration with Distributed Arrays on the World&rsquo\; s Largest Interactive AI Supercomputer</h4>\n<h4>Jeremy Kepner</h4>\n<p>&n bsp\;</p>\n<h4>Please register in advance for this seminar even if you pla n to attend in person at</h4>\n<p class="p1">&lt\;https://acm-org.zoom.us/ webinar/register/1617607374551/WN_lYs4lxKfSlGkMVq71ibN-g&gt\;</p>\n<p>Afte r registering\, you will receive a confirmation email containing informati on about joining the webinar.</p>\n<p>Indicate on the registration form if you plan to attend in person. This will help us determine whether the roo m is close to reaching capacity. We plan to serve light refreshments (prob ably pizza) before the talk starting at around 6:30 pm. Letting us know yo u will come in person will help us determine how much pizza to order.</p>\ n<p>We may make some auxiliary material such as slides and access to the r ecording available after the seminar to people who have registered.</p>\nA bstract:\n<p>High level programming languages and GPU accelerators are pow erful enablers for a wide range of applications. Achieving scalable vertic al (within a compute node)\, horizontal (across compute nodes)\, and tempo ral (over different generations of hardware) performance while retaining p roductivity requires effective abstractions. Distributed arrays are one su ch abstraction that enables high level programming to achieve highly scala ble performance. Distributed arrays achieve this performance by deriving p arallelism from data locality\, which naturally leads to high memory bandw idth efficiency. This talk explores distributed array performance on a var iety of hardware. Scalable performance is demonstrated within and across C PU cores\, CPU nodes\, and GPU nodes. The interactive AI supercomputing ha rdware used spans decades and allows a direct comparison of hardware impro vements over this time range.</p>\n<p>&nbsp\;</p>\n<p>Bio:</p>\n<p>Dr. Jer emy Kepner is an MIT Lincoln Laboratory Fellow. He founded the Lincoln Lab oratory Supercomputing Center and pioneered the establishment of the Massa chusetts Green High Performance Computing Center. He has developed novel b ig data and parallel computing software used by thousands of scientists an d engineers worldwide. He has led several embedded computing efforts\, whi ch earned him a 2011 R&amp\;D 100 Award. Kepner has chaired the SIAM Data Mining conference\, the IEEE Big Data conference\, and the IEEE High Perfo rmance Extreme Computing conference. Kepner is the author of two bestselli ng books\, Parallel MATLAB for Multicore and Multinode Computers\, and Gra ph Algorithms in the Language of Linear Algebra. His peer-reviewed publica tions include works on abstract algebra\, astronomy\, astrophysics\, cloud computing\, cybersecurity\, data mining\, databases\, graph algorithms\, health sciences\, plasma physics\, signal processing\, and 3D visualizatio n. In 2014\, he received Lincoln Laboratory's Technical Excellence Award.< /p>\n<p>Kepner holds a BA degree in astrophysics from Pomona College and a PhD degree in astrophysics from Princeton University. He is a fellow of t he Society of Industrial Applied Mathematics (SIAM) and is a faculty advis or to the MIT SIAM student group.</p>\n<p>Directions to 32-G449 - MIT Stat a Center\, 32 Vassar Street\, Cambridge\, MA: Please use the main entrance to the Stata Center at 32 Vassar Street (the entrance closest to Main str eet) as those doors will be unlocked. Upon entering\, proceed to the eleva tors which will be on the right after passing a large set of stairs and a MITAC kiosk. Take the elevator to the 4th floor and turn right\, following the hall to an open area\; 32-G449 will be on the left.&nbsp\;<a href="ht tps://whereis.mit.edu/?go=32">Location of Stata on campus map</a></p>\n<p> &nbsp\;</p>\n<p class="norm10">This joint meeting of the Boston Chapter of the IEEE Computer Society and GBC/ACM will be hybrid (in person and onlin e).</p>\n<p>Up-to-date information about this and other talks is available online at https://ewh.ieee.org/r1/boston/computer/. You can sign up to re ceive updated status information about this talk and informational emails about future talks at https://mailman.mit.edu/mailman/listinfo/ieee-cs\, o ur self-administered mailing list.</p>\n</blockquote> END:VEVENT END:VCALENDAR