Nirmesh Malviya

Nirmesh Malviya
Photo: Jason Dorfman, CSAIL photographer

Where did you grow up:

I grew up in Indore, a major city in Central India.

What was your academic path before coming to grad school at MIT?

I have a Bachelor's in Computer Science and Engineering from one of the Indian Institutes of Technology - IIT Kanpur.

What did you want to be when you were younger? Is that still an interest of yours?

When I was very young, I aspired to play cricket professionally for India. By middle school, I realized I was better at math and cricket became more of a hobby. Even though I have switched to playing individual sports, such as badminton and tennis over the years, I still follow cricket actively.

What is something most people would be surprised to learn about you?

Most people familiar with me know I love listening to all kinds of music, but I happen to enjoy playing music way more than I like listening to it. Learning to make my guitar cry and sing is one of my favorite pastimes.

What department are you currently working in, and when did you start there?

I started grad school in fall 2010. I am in EECS.

What are you working on and why are you passionate about it?

I am currently working on recovery techniques for in-memory relational databases. It's extremely exciting because transactional web workloads running on our system see orders of magnitude performance improvement and the system can scale up almost linearly owing to our in-memory lockless database architecture. Durability is an issue, however, and I have been looking at different strategies to keep the impact of recovery on system performance minimal.

What is your favorite thing about working at CSAIL?

There are a lot of smart people in CSAIL, and they are more than happy to tell you that you are wrong! Seriously speaking, bouncing your ideas off of such smart peers helps immensely in the early stages of research. CSAIL's also very diverse, and it is extremely rewarding to talk to people who are working on an altogether different set of research problems in computer science. Then there is the Stata Center, it's fun finding your way around this unique building.

What effect do you think your area of work will have on the world in the next decade?

Computing hardware is evolving rapidly: processors are becoming massively multi-core, solid state disks are starting to become commonplace and the age of non-volatile main memory such as PCME is not far. Yet all big data problems out there are being solved using either traditional database systems designed forty years ago for single core processors with volatile main memories, or by writing ad-hoc applications written for specific use cases. My work is a step in the direction of research needed to rethink the architecture of database systems from the ground up. Data is proliferating at the Internet scale; analyzing and effectively managing it will only become increasingly harder over the next decade unless an attempt to solve the problem systematically is made.

What are your future plans?

I would be interested in working in the tech industry. Both my summer stints in industry, one at Google and the other at Microsoft Research, were fun experiences and I got to work on interesting real problems. I might also do a startup down the road, but only if I am particularly passionate about some idea, rather than for the sake of it.

What advice would you give a prospective CSAIL graduate student?

I would say keep an open mind about what you want to do as well as what problems interest you. Don't restrict yourself to your area of research alone and build up a breadth of knowledge by taking classes or doing research across different areas of computer science. You might end up discovering another area you find even more interesting or something which has direct applications to problems in your area of research.