- Research Group
My research spans a number of areas in computer systems, including distributed systems, large-scale/cloud computing and computer networking with emphasis on programmable networks.
At KAUST, I lead the SANDS Lab.
Here is a representative subset of projects I am working on:
A general framework for implementing, studying and evaluating gradient compression techniques for distributed deep learning.
We accelerate distributed deep learning by designing an in-network aggregation primitive using a programmable switch dataplane.
DAIET: Data Aggregation In nETwork
We propose the first in-network aggregation primitive based on programmable network switches.
In my past, I have been active in SDN and programmable networks. Major projects included:
- Dynam-IX: Dynamic Interconnection eXchange
- SIXPACK: Securing Internet eXchange Points Against Curious onlooKers
- NetPaxos: Network support for consensus protocols
- Software Transactional Networking
- Panopticon: Incremental SDN Deployment in Enterprise Networks
I am privileged to be working with these very talented individuals:
- M. Bilal (UCLouvain and IST, co-advised with Rodrigo Rodrigues)
- Arnaud Dethise
- Jiawei Fei
- Chen-Yu (Elton) Ho
- Waleed Reda (UCLouvain and KTH, co-advised with Dejan Kostic)
- Atal Sahu
- Fatimah Zohra
- Lalith Suresh (2016, now Researcher at VMware Research)
- Dan Levin (2014, co-founder and CEO of Stacktile GmbH)
Research MS students (while at KAUST)
- Hassan Alsibyani (2018, now Technical Lead at Wasphi)
- Yousef Alowayed (2018, now Software Engineer at Hellosuper)