Amitabha Roy

I am a research scientist at Intel LabsMy interests span systems, computer architecture and algorithms. I enjoy studying, building and tuning systems to solve challenging problems. Some of the interesting things I've done are building graph processing systems that tackle trillions of edges, developed a completely transparent software transactional memory implementation for x86 machine code and developed a parallel algorithm for verifying the x86 memory consistency model that is used by Intel to verify their multicore products before they are shipped to customers. My current work at Intel Labs is in hardware software co-design that enables Intel's silicon to be used in a variety of interesting ways in contemporary datacenters, web services and HPC environments.    

Contact: 

amitabha dot roy at gmail dot com 

amitabha dot roy at intel dot com

Twitter

Linkedin           

Publications [Google Scholar  DBLP]


Subramanya R Dulloor, Amitabha Roy, Zheguang Zhao, Narayanan Sundaram, Nadathur Satish, Rajesh Sankaran, Jeff Jackson, Karsten Schwan.

Amitabha Roy, Laurent Bindschaedler, Jasmina Malicevic, Willy Zwaenepoel

Jiaqing Du, Amitabha Roy, Calin Iorgulescu, Willy Zwaenepoel.

Karthik Nilakant, Valentin Dalibard, Amitabha Roy, Eiko Yoneki.

Amitabha Roy, Ivo Mihailovic, Willy Zwaenepoel. 

Jiaqing Du, Sameh Elnikety, Amitabha Roy, Willy Zwaenepoel. 

Amitabha Roy, Steven Hand, Tim Harris. 
Weak Atomicity for the x86 Memory Consistency Model, Journal of Parallel and Distributed Computing, 72(2012) [preprint]

Amitabha Roy, Steven Hand, Tim Harris. 

Amitabha Roy, Steven Hand, Tim Harris. 

Amitabha Roy, Stephan Zeisset, Charles J. Fleckenstein,John C. Huang. 

Amitabha Roy, K. Gopinath. 

Short Papers/Early Ideas


Jiaqing Du, Calin Iorgulescu, Amitabha Roy, Willy Zwaenepoel. 

Jasmina Malicevic, Amitabha Roy, Willy Zwaenepoel. 

Amitabha Roy, Timothy Jones. ALLARM: Optimizing Sparse Directories for Thread-Local Data, DATE 2014

Eiko Yoneki, Amitabha Roy. Scale-up Graph Processing: A Storage-centric View, GRADES 2013

Amitabha Roy, Steven Hand, Tim Harris. Exploring the Limits of Disjoint Access Parallelism, HotPar 2009

Technical Reports

Eiko Yoneki, Amitabha Roy, Derek Murray.

Amitabha Roy, Karthik Nilakant, Valentin Dalibard, Eiko Yoneki. 

Eiko Yoneki, Amitabha Roy. A Unified Graph Query Layer for Multiple Databases, UCAM-CL-TR 2012

Amitabha Roy. Memory Hierarchy Sensitive Graph Layout, Arxiv 2012


Community Service 


Sponsorship co-chair for Eurosys 2017.
PC member for HPGP 2016.
PC member for Eurosys 2015
Co-organized a Dagstuhl workshop on "Systems and Algorithms for Large Scale Graph Analytics".
Invited talk on the Post-Doctoral experience at EuroDW2014.

 
Past Associations

[2012-2015] Post-doc at LABOS in EPFL, where I was primarily responsible for the X-Stream and Chaos graph processing systems and contributed to the Orbe geo-replicated key value storage system.

[2011-2012] Post-doc in the Computer Architecture Group at the University of Cambridge Computer Laboratory, where I worked on a technique to adapt directory controllers for thread-private data with Timothy Jones. Simultaneously I also worked with Eiko Yoneki in the Systems Research Group on ways to mitigate IO stalls when processing large graphs - a collaboration that later became the Graphcam project.

[2011] Short stint at Acunu where I worked on analysing the performance of their filesystem.
 
[2007-2011] PhD student in the Networks and Operating Systems (NetOS) research group at the Computer Laboratory in the University of Cambridge. I was supervised by Steven Hand and Tim Harris. My PhD thesis, titled "Software lock elision for x86 machine code", argues for separation of mechanism and policy in the context of software transactional memory. It presents the design and implementation of a system that can be used to automatically elide legacy locks in x86 machine code. Transactional memory therefore becomes an optional mechanism for synchronization, with legacy locks or atomic blocks implemented using legacy locks being used to specify synchronization policy. A talk I've given describes many of the basic ideas.

[2004-2007] Design Engineer at Intel where I worked for some time on memory consistency verification and later did performance modeling and analysis of CPUs. Among other things, I designed a parallel algorithm to verify memory consistency test results using graphs, which is widely used in post-silicon validation at Intel. More work related things can be found in my linkedin profile.

[2002-2004] Master's degree student in computer science from the Computer Science department at IISc Bangalore. While there, I worked in the Computer Architecture and Systems Laboratory under Prof. K. Gopinath on probabilistic timed automata for 802.11 networks.
 
[1998-2002] Undergraduate engineering student at Jadavpur University in the Department of Computer Science and Engineering.