BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260724T151408Z
LOCATION:Bldg. 6 - 102
DTSTART;TZID=Europe/Stockholm:20260629T133000
DTEND;TZID=Europe/Stockholm:20260629T153000
UID:submissions.pasc-conference.org_PASC26_sess106@linklings.com
SUMMARY:MS1D - 2nd Sustainable Computing for AI and Data-Intensive Infrast
 ructures
DESCRIPTION:Access the recording\n\nOrganizer(s): Denisa-Andreea Constanti
 nescu (EPFL), Florina Ciorba (University of Basel), and Can Hankendi (Bost
 on University)\n\nArtificial Intelligence (AI) has become a primary driver
  of compute demand and energy consumption, with workloads such as large la
 nguage models and AI-powered computational science applications requiring 
 sustained, infrastructure-scale resources. At the same time, modern data c
 enters operate under increasingly tight power, carbon, and power grid cons
 traints, particularly as they integrate variable renewable energy sources.
  Together, these trends highlight the growing tension between performance 
 scalability, energy availability, and environmental impact. Addressing the
 se challenges requires integrated efforts from component-level optimizatio
 n to co-designing computing infrastructures, workloads, and energy systems
 . This minisymposium examines how sustainability can be embedded by design
  across AI and data-intensive computing infrastructures. Topics include (1
 ) datacenter-power grid co-design to enable load flexibility under renewab
 le integration, (2) infrastructure-level optimization of AI and AI-powered
  workloads for improved energy efficiency, (3) hardware-software co-design
  approaches that explore accelerator design trade-offs for AI and scientif
 ic computing, and (4) methods, metrics, and lifecycle assessment tools. Th
 e session also highlights the economic and environmental tensions between 
 cost-optimal and emissions-optimal system configurations. By bringing toge
 ther researchers and practitioners working on architectures, systems, and 
 infrastructure design, this session aims to foster a shared understanding 
 of the design principles needed to support sustainable growth in AI and da
 ta-intensive computing.\n\nModelling the Environmental Footprint of Future
  AI and HPC Datacenters: Enabling Hardware-Software Codesign Including Lif
 e Cycle Assessment\n\nArtificial intelligence (AI) has emerged as a pivota
 l force shaping the future of scientific discovery and technological innov
 ation in day-to-day life. It has become a key enabler for future scientifi
 c research and development of new applications. As data centers dedicated 
 to these workloads experie...\n\n\nNitish Satya Murthy, Vincent Schelleken
 s, and Timon Evenblij (IMEC)\n---------------------\nTowards Green AI: Red
 ucing the Environmental Footprint of LLM Workloads\n\nContemporary large-s
 cale computing systems are becoming increasingly heterogeneous, complex, a
 nd decentralized, driven by growing digitization demands. This transformat
 ion is most evident in the rapid rise of Generative AI applications, which
  have fueled the expansion of hyperscale data centers, now...\n\n\nShashik
 ant Ilager (University of Amsterdam) and Ivona Brandic (TU Wien)\n--------
 -------------\nHow Design Choices Shape Sustainable Computing: A Hardware 
 Accelerator Perspective\n\nData centers are experiencing rapidly rising en
 ergy consumption and associated greenhouse-gas emissions. Today, GPUs are 
 the dominant accelerator in many data centers, yet they are not necessaril
 y the most efficient choice for every application domain because they rema
 in relatively general-purpose. ...\n\n\nRuben Rodriguez Alvarez and Denisa
 -Andreea Constantinescu (EPFL)\n---------------------\nMiddlebox: Resolvin
 g the Datacenter-Grid Tension and Decarbonizing Both\n\nDatacenter’s (DCs)
  constant power requirement is difficult to balance with variable solar an
 d wind generation.  DC load flexibility is the key to solving the problem,
  but it conflicts stable capacity requirements.\n\nWe propose ``power Midd
 lebox'', a new system architecture, as a bridge. Middleb...\n\n\nAndrew Ch
 ien (Univ of Chicago, Argonne National Laboratory)\n\nDomain: Engineering,
  Computational Methods and Applied Mathematics\n\nSession Chairs: Denisa-A
 ndreea Constantinescu (EPFL) and Florina Ciorba (University of Basel)
END:VEVENT
END:VCALENDAR