This program is tentative and subject to change.

The carbon and water footprint of large-scale computing systems poses serious environmental sustainability risks. In this study, we discover that, unfortunately, carbon and water sustainability are at odds with each other - and, optimizing one alone hurts the other. Toward that goal, we introduce, WaterWise, a novel job scheduler for parallel workloads that intelligently co-optimizes carbon and water footprint to improve the sustainability of geographically distributed data centers.

This program is tentative and subject to change.

Tue 4 Mar

Displayed time zone: Pacific Time (US & Canada) change

11:20 - 12:20
Session 7: Scheduling and Resource Management (Session Chair: TBA)Main Conference at Acacia D
11:20
20m
Talk
SGDRC: Software-Defined Dynamic Resource Control for Concurrent DNN Inference on NVIDIA GPUs
Main Conference
Yongkang Zhang HKUST, Haoxuan Yu HKUST, Chenxia Han CUHK, Cheng Wang Alibaba Group, Baotong Lu Microsoft Research, Yunzhe Li Shanghai Jiao Tong University, Zhifeng Jiang HKUST, Yang Li China University of Geosciences, Xiaowen Chu Data Science and Analytics Thrust, HKUST(GZ), Huaicheng Li Virginia Tech
11:40
20m
Talk
DORADD: Deterministic Parallel Execution in the Era of Microsecond-Scale Computing
Main Conference
Scofield Liu Imperial College London, Musa Unal EPFL, Matthew J. Parkinson Microsoft Azure Research, Marios Kogias Imperial College London; Microsoft Research
12:00
20m
Talk
WaterWise: Co-optimizing Carbon- and Water-Footprint Toward Environmentally Sustainable Cloud Computing
Main Conference
Yankai Jiang Northeastern University, Rohan Basu Roy Northeastern University, Raghavendra Kanakagiri Indian Institute of Technology Tirupati, Devesh Tiwari Northeastern University