Purdue University - Department of Computer Science - Akshay Jajoo Skip to main content
Akshay Jajoo photo

Akshay Jajoo

Graduate Student

Joined department: Fall 2015


Bachelor of Technology, Indian Institute of Technology, Computer Science (2015)

Akshay is pursuing Ph.D. in Computer Science. He works at DSN lab, under Dr. Y.C Hu's guidance. He completed his Bachelor of Technology in Computer Science and Engineering from IIT Guwahati in 2015. He was awarded Dr. Shankar Dayal Sharma gold medal at IITG for excellence in academics and extra curricular activities.

Selected Publications

Graviton: Twisting space and time to speedup coflows, HotCloud 2016

Authors: Akshay Jajoo, Rohan Gandhi, Y. Charlie Hu

CoFlow is a networking abstraction. A group of flows sharing common end goal are termed as CoFlow. Graviton provides proof of the concept that by taking spatial dimension of CoFlows into account their average completion time can be improved significantly. We have shown that width, number of different ports involved, is a very strong and significant indicator of optimal CoFlow ordering. CoFlow scheduling is an NP-Hard problem. On a high level, the proposed heuristic sorts CoFlows according to their width. We have achieved a speedup of 1.25x(P50) and 8.0x(P90) in CoFlow completion time as compared to the existing schedulers.

Saath: Speeding up CoFlows by Exploiting the Spatial Dimension, CoNext 2017

Authors: Akshay Jajoo, Rohan Gandhi, Y. Charlie Hu, Cheng-Kok Koh

Prior CoFlow schedulers approximate the classic online SJF scheduling by using a global scheduler to sort CoFlows into multiple priority queues, and a local scheduler at each network port to schedule the flows of CoFlows using FIFO. Such a division of the scheduling suffers 2 problems:

  1. The flows of a CoFlow may suffer the out-of-sync problem negatively affecting the CoFlow completion time (CCT);
  2. FIFO scheduling of flows at each port bears no notion of SJF, leading to suboptimal CCT.

Saath is an online CoFlow scheduler that overcomes the above drawbacks by explicitly exploiting the spatial dimension of CoFlows. Our evaluation using an Azure testbed and simulations of production cluster traces show that compared to Aalo, Saath reduces the CCT on average by 1.78x (P90 = 4.50x), which reduces the job completion time on average by 1.46x (P90 = 1.86x).

Your Coflow has Many Flows: Sampling them for Fun and Speed (Philae), USENIX ATC 2019

Authors: Akshay Jajoo, Y. Charlie Hu, Xiaojun Lin

Coflow scheduling improves data-intensive application performance by improving their networking performance. State-of-the-art online coflow schedulers in essence approximate the classic Shortest-Job-First (SJF) scheduling by learning the coflow size online. In particular, they use multiple priority queues to simultaneously accomplish two goals: to sieve long coflows from short coflows, and to schedule short coflows with high priorities. Such a mechanism pays high overhead in learning the coflow size: moving a large coflow across the queues delays small and other large coflows, and moving similar-sized coflows across the queues results in inadvertent round-robin scheduling

We propose Philae, a new online coflow scheduler that exploits the spatial dimension of coflows, i.e., a coflow has many flows, to drastically reduce the overhead of coflow size learning. Philae pre-schedules sampled flows of each coflow and uses their sizes to estimate the average flow size of the coflow. It then resorts to Shortest Coflow First, where the notion of shortest is determined using the learned coflow sizes and coflow contention. We show that the sampling-based learning is robust to flow size skew and has the added benefit of much improved scalability from reduced coordinator-local agent interactions. Our evaluation using an Azure testbed, a publicly available production cluster trace from Facebook shows that compared to the prior art Aalo, Philae reduces the coflow completion time (CCT) in average (P90) cases by 1.50x (8.00x) on a 150-node testbed and 2.72x (9.78x) on a 900-node testbed. Evaluation using additional traces further demonstrates Philae's robustness to flow size skew.

Last Updated: Dec 4, 2020 2:02 PM

Department of Computer Science, 305 N. University Street, West Lafayette, IN 47907

Phone: (765) 494-6010 • Fax: (765) 494-0739

Copyright © 2020 Purdue University | An equal access/equal opportunity university | Copyright Complaints

Trouble with this page? Disability-related accessibility issue? Please contact the College of Science.