BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T143034Z
LOCATION:B312
DTSTART;TZID=America/New_York:20241117T090000
DTEND;TZID=America/New_York:20241117T173000
UID:submissions.supercomputing.org_SC24_sess739@linklings.com
SUMMARY:Sustainable Supercomputing
DESCRIPTION:Sustainable supercomputing is a pressing topic for our communi
 ty, industry, and governments. Supercomputing has an ever-increasing need 
 for computational cycles while facing increasing challenges of delivering 
 performance/Watt advances and within the context of climate change, the dr
 ive towards net-zero, and geo-political-economic pressures. Improving supe
 rcomputing sustainability provides many opportunities considering an end-t
 o-end, holistic view of the HPC system, facility, site, and broader enviro
 nment. All elements of the HPC system must be considered; from low-level c
 ircuits, up the software stack and beyond to power/cooling systems. The dr
 ive towards more sustainable supercomputing requires measurements, metrics
 , goals, and improvement processes.\n\nhws: A Tool for Monitoring Hardware
  Metrics Across Diverse Vendors: A Case Study on Hyperparameter Optimizati
 on Algorithms\n\nDue to modern hardware's constantly growing energy demand
 s, it is important to consider energy efficiency and power consumption. Es
 pecially in the age of AI, where a massive amount of computational power i
 s necessary, energy consumption and the costs involved can become a signif
 icant problem. Howeve...\n\n\nMarcel Breyer, Alexander Van Craen, Peter Do
 manski, and Dirk Pflüger (University of Stuttgart, Germany; University of 
 Stuttgart, Institute for Parallel and Distributed Systems)\n--------------
 -------\nCEEMS: A Resource Manager Agnostic Energy and Emissions Monitorin
 g Stack\n\nCompute Energy & Emissions Monitoring Stack (CEEMS) has been de
 signed to report energy usage of compute workloads in real time for HPC an
 d cloud platforms alike. Besides CPU energy usage, it supports reporting e
 nergy usage of workloads on NVIDIA and AMD GPU accelerators. CEEMS has bee
 n built around ...\n\n\nMahendra Paipuri (CNRS)\n---------------------\nSu
 stainable Supercomputing — Lunch Break\n---------------------\nSustainable
  Supercomputing — Morning Break\n---------------------\nSustainable AI: Ex
 periences, Challenges and Recommendations\n\nThe use of Artificial Intelli
 gence (AI) and Machine Learning (ML) as part of scientific workloads is be
 coming increasingly widespread. It is imperative to understand how to conf
 igure AI and ML applications on HPC systems to optimise their performance 
 and energy efficiency, thereby minimising their e...\n\n\nEleanor Broadway
 , Joseph Lee, and Michele Weiland (Edinburgh Parallel Computing Centre (EP
 CC))\n---------------------\nAIOps and Sustainability: Transforming Data C
 enters for a Greener Future\n\nEnterprise and high-performance computing d
 ata centers are dealing with thousands of sensor metrics and associated da
 ta. A top-end target for exascale machines is 10 million data points per s
 econd. The escalating volume and speed of data generation are making thing
 s more difficult, and outages are i...\n\n\nSubrahmanya Vinayak Joshi, Ser
 gey Serebryakov, Deepak Nanjundaiah, Tejas Hegde, and Martin Foltin (Hewle
 tt Packard Enterprise (HPE))\n---------------------\nExploring the Frontie
 rs of Energy Efficiency using Power Management at System Scale\n\nIn the f
 ace of surging power demands for exascale HPC systems, this work tackles t
 he critical challenge of understanding the impact of software-driven power
  management\ntechniques like Dynamic Voltage and Frequency Scaling (DVFS) 
 and Power Capping. These techniques have been actively developed over t...
 \n\n\nAhmad Maroof Karimi, Matthias Maiterth, Woong Shin, Naw Safrin Satta
 r, Hao Lu, and Feiyi Wang (Oak Ridge National Laboratory (ORNL))\n--------
 -------------\nEE-HPC a Framework for Energy Efficient HPC System Manageme
 nt\n\nThe energy consumption has become a major cost\nfactor in the procur
 ement and operation of large scale HPC data\ncenters. In addition, funding
  bodies and governments are starting\nto focus on assessment and improveme
 nt of energy efficiency, as\nwell as reducing the overall environmental im
 pact of data c...\n\n\nDavid Brayford (Hewlett Packard Labs); Radita Liem 
 (RWTH Aachen University); and Thomas Gruber (Friedrich-Alexander Universit
 y, Erlangen-Nuremberg)\n---------------------\nTowards Sustainable Post-Ex
 ascale Leadership Computing\n\nAs computing systems approach the limits of
  traditional silicon technology, the diminishing returns in performance pe
 r watt present a significant barrier to sustaining growth in HPC. From a l
 arge-scale scientific supercomputing facility point of view, we propose a 
 multifaceted strategy toward specia...\n\n\nWoong Shin, James B. White III
 , Wael Elwasif, Rafael Ferreira da Silva, Christopher Zimmer, Bronson Mess
 er, Reuben Budiardja, Antigoni Georgiadou, Verónica Melesse Vergara, Jack 
 Lange, Matthias Maiterth, Tim Osborne, Leah Huk, John Holmen, Nick Hagerty
 , Ahmad Maroof Karimi, Thomas Naughton, Ryan Adamson, Ryan Prout, Feiyi Wa
 ng, Scott Atchley, Kevin G. Thach, Thomas Beck, and Sarp Oral (Oak Ridge N
 ational Laboratory (ORNL))\n---------------------\nPanel/Wrap up\n\nMichae
 l Woodacre (HPE)\n---------------------\nSustainable Supercomputing — Afte
 rnoon Break\n---------------------\nIncreasing Energy Efficiency of Astrop
 hysics Simulations Through GPU Frequency Scaling\n\nThe growing demand for
  HPC necessitates significant energy consumption, posing a sustainability 
 challenge for HPC centers, users, and society, especially\ndue to stricter
  environmental regulations. While efforts exist to reduce overall system e
 nergy consumption, optimizations\nfor GPU-based workloads,...\n\n\nOsman S
 eckin Simsek (University of Basel, Switzerland); Jean-Guillaume Piccinali 
 (Swiss National Supercomputing Centre (CSCS)); and Florina Ciorba (Univers
 ity of Basel, Switzerland)\n---------------------\nInput-Dependent Power U
 sage in GPUs\n\nGPUs are known to be power-hungry, and due to the boom in 
 artificial intelligence, they are the major contributors to the high power
  demands of datacenters. Most GPU usage in these popular workloads consist
 s of large general matrix-matrix multiplications (GEMMs), which have there
 fore been optimized ...\n\n\nTheo Gregersen (University of Washington, Car
 negie Mellon University); Pratyush Patel (University of Washington); and E
 sha Choukse (Microsoft Corporation)\n---------------------\nAnalysis of Po
 wer Consumption and GPU Power Capping for MILC\n\nPower has been a key con
 straint for supercomputers, and limitations on power become increasingly n
 oticeable through the exascale era. Limited power availability pushes the 
 facilities to operate under power constraints and develop power management
  methods, making it crucial to understand applications...\n\n\nFatih Acun 
 (Boston University), Zhengji Zhao and Brian Austin (Lawrence Berkeley Nati
 onal Laboratory (LBNL)), Ayse Coskun (Boston University), and Nicholas J. 
 Wright (Lawrence Berkeley National Laboratory (LBNL))\n-------------------
 --\nPawsey's Sustainability Journey\n\nMaciej Cytowski (Pawsey Supercomput
 ing Research Centre)\n---------------------\nNavigating Exascale Operation
 al Data Analytics: From Inundation to Insight\n\nIn this paper, we address
  the challenges in achieving sustainable data-driven efficiency by providi
 ng a detailed exploration of the end-to-end operational data analytics (OD
 A) framework that evolved through two generations of supercomputer systems
  at the Oak Ridge Leadership Computing Facility (OLCF...\n\n\nWoong Shin, 
 Tim Osborne, Ahmad Maroof Karimi, Rachel Palumbo, Alex May, Corwin Lester,
  Jesse Hines, Naw Safrin Sattar, Leah Huk, Scott Simmerman, Wesley Brewer,
  Jeffrey Miller, Ryan Adamson, Olga Kuchar, Ryan Prout, Feiyi Wang, Scott 
 Atchley, and Sarp Oral (Oak Ridge National Laboratory (ORNL))\n-----------
 ----------\nVendor-neutral and Production-grade Job Power Management in Hi
 gh Performance Computing\n\nPower management and energy efficiency are cri
 tical\nresearch areas for exascale computing and beyond, necessitating\nre
 liable telemetry and control for distributed systems. Despite this\nneed, 
 existing approaches present several limitations precluding\ntheir adoption
  in production. These limitations in...\n\n\nNaman Kulshresthha (Clemson U
 niversity); Tapasya Patki, Jim Garlick, and Mark Grondona (Lawrence Liverm
 ore National Laboratory (LLNL)); and Rong Ge (Clemson University)\n\nTag: 
 Energy Efficiency, HPC Infrastructure, Sustainability\n\nRegistration Cate
 gory: Workshop Reg Pass\n\nSession Chairs: Cate Berard (US Department of E
 nergy); James H. Rogers (Oak Ridge National Laboratory (ORNL)); Fumiyoshi 
 Shoji (RIKEN Center for Computational Science (R-CCS), Center for Computat
 ional Science); Michèle Weiland (EPCC, The University of Edinburgh; The Un
 iversity of Edinburgh); and Mike Woodacre (Hewlett Packard Enterprise (HPE
 ))
END:VEVENT
END:VCALENDAR