BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Asia/Kolkata
X-WR-TIMEZONE:Asia/Kolkata
BEGIN:VEVENT
UID:173@cds.iisc.ac.in
DTSTART;TZID=Asia/Kolkata:20260107T143000
DTEND;TZID=Asia/Kolkata:20260107T153000
DTSTAMP:20260105T165417Z
URL:https://cds.iisc.ac.in/events/cloud-seminar-cds-102-january-07-1430-to
 wards-sustainable-ai-optimizing-energy-and-carbon-footprints-in-large-scal
 e-llm-workloads/
SUMMARY:{Cloud Seminar} @ CDS: #102\, January 07\, 14:30: "Towards Sustaina
 ble AI: Optimizing Energy and Carbon Footprints in Large-Scale LLM Workloa
 ds"
DESCRIPTION:\n\nCLOUD COMPUTING SEMINAR SERIES\n\n\n\nTitle: Towards Sustai
 nable AI: Optimizing Energy and Carbon Footprints in Large-Scale LLM Workl
 oads\nSpeaker: Shashikant Ilager\, University of Amsterdam\, the Netherlan
 ds\nDate &amp\; Time : January 07\, 2026 (Wednesday)\, 14:30\nVenue: CDS 1
 02 Seminar Hall\n\n\n\nABSTRACT\n\nContemporary large-scale computing syst
 ems are becoming increasingly heterogeneous\, complex\, and decentralized\
 , driven by the growing demand for digitization. This transformation is mo
 st evident in the rapid rise of Generative AI applications\, which have fu
 eled the expansion of hyperscale data centers\, now often referred to as A
 I factories. While these infrastructures enable unprecedented computationa
 l capabilities\, they also impose high environmental costs by contributing
  heavily to global energy consumption and carbon emissions. Assessing and 
 managing resource and energy use in such distributed systems is a complex 
 challenge. It requires addressing the heterogeneity of hardware platforms 
 and their diverse power characteristics\, the wide range of user workloads
  and software stacks\, and the risk of performance degradation when energy
 -saving measures are applied. In this talk\, I will present our recent res
 earch on data-driven modeling and optimization techniques for energy-aware
  management of large-scale computing systems. In particular\, I will discu
 ss how Large Language Model (LLM)-based applications can be fine-tuned thr
 ough hardware and software co-design strategies to balance performance wit
 h sustainability objectives.\n\nBIOGRAPHY\n\nShashikant Ilager is an Assis
 tant Professor at the Informatics Institute\, University of Amsterdam (UvA
 )\, where he is a member of the Multiscale Networked Systems Research Grou
 p. Before joining the UvA\, he worked as a postdoctoral researcher in the 
 High Performance Computing group at the Vienna University of Technology (T
 U Wien)\, Austria. He was a visiting researcher at INRIA\, IMT Atlantique 
 in 2023\, and at IBM Research\, Yorktown Heights\, USA\, in 2024. He compl
 eted his PhD in Computer Science and Engineering at the CLOUDS Lab\, Unive
 rsity of Melbourne (UniMelb)\, Australia\, in 2021. He is currently the co
 ordinator of the European Horizon project GreenDIIGT\, leading a consortiu
 m of 15 European research institutes and organizations that are developing
  policies\, tools\, and innovative solutions for green research infrastruc
 tures. His research lies at the intersection of distributed systems\, ener
 gy efficiency\, and machine learning. He focuses on performance engineerin
 g\, data-driven optimization for energy efficiency\, and sustainable AI. H
 is teaching and research have been recognized with several awards\, includ
 ing the Excellence in Teaching Award at UniMelb in 2018\, the IEEE TCCLD O
 utstanding PhD Thesis Award in 2021\, and Best Paper Awards at CCGRID 2020
  and UCC 2023. Please visit his homepage for more details: https://shashik
 antilager.com/\n\nHost Faculty: Prof. Yogesh Simmhan\n\n\n\nABOUT: The IBM
 -IISc Hybrid Cloud Lab (IIHCL) hosted at IISc is curating the Cloud Comput
 ing Seminar series with guest speakers from Industry and Academia speaking
  about the latest technologies and research on Cloud and edge computing\, 
 distributed computing systems\, and AI/ML/Big Data platforms.\n\n\n\nALL A
 RE WELCOME
CATEGORIES:Events,Talks
END:VEVENT
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
X-LIC-LOCATION:Asia/Kolkata
BEGIN:STANDARD
DTSTART:20250107T143000
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR