Akshay Jajoo is a senior researcher with the Decentralized Networks and Systems Research group at Bell Labs. While Akshay's current research focuses on the Decentralized Web, he is also interested in scheduling and online learning for distributed jobs and cloud computing. Akshay got his Ph.D. in Computer Science with a specialization in Distributed Systems from Purdue University. He has published his Ph.D. work in top venues like USENIX NSDI, IEEE ToN, USENIX ATC, and ACM CoNEXT. Before this, he got a B.Tech from IIT Guwahati in Computer Science and Engineering, where he was also recognized as the best all-rounder of the institute. Akshay enjoys swimming, playing Tabla, and strolling in the woods for recreational purposes. Akshay also serves as a Mentor on University courses and Ph.D. committees.
Ph.D. in Computer Science, Purdue University, USA , 2020
B.Tech in Computer Science and Engineering, Indian Institute of Technology (IIT) Guwahati, 2015
TPC member and Reviewer for IEEE ICNP, IFIP Networking, IEEE ToN, and IEEE ToCC.
Selected articles and publications
SLearn (A Case for Task Sampling based Learning for Cluster Job Scheduling), USENIX NSDI 2022
Deep Learning-based Channel State Information Prediction with Incomplete History, IEEE WCNC 2022
A Case for Sampling-Based Learning Techniques in Coflow Scheduling, IEEE ToN 2021
Philae (Your Coflow has Many Flows: Sampling them for Fun and Speed), USENIX ATC 2019
Saath: Speeding up CoFlows by Exploiting the Spatial Dimension, ACM CoNEXT 2017
Graviton: Twisting space and time to speedup CoFlows, USENIX HotCloud 2016
Honors & Awards
DAAD Fellowship, Charpak Fellowship