amar phanishayee

Intro | Publications | Misc.

Hi !

As of May 2024, I am a Research Scientist at Meta. For close to 12 years prior to that, I worked at the creative haven of Microsoft Research in Redmond, where I was a Sr. Principle Researcher. There, among other things, I started and led Project Fiddle, where my collaborators and I worked on systems for large-scale DNN training and serving. I was also fortunate, working with a group of fearless engineers, to be one of a small group of software-systems architects for Microsoft's early forays in 1P AI Supercomputing efforts.

Earlier still, many moons ago, I got my PhD in Computer Science at Carnegie Mellon University where I worked with Dave Andersen. The core of my work, with fellow collaborators then, anchored around: (i) FAWN (energy-efficient distributed systems, especially for large-scale random-access IO-bound workloads), and (ii) Incast (catastrophic TCP throughput collapse in datacenter-scale workloads). Incast has since become an adjective and Wimpy is used in straight-faced technical conversations; who'd a thunk it?

The goal of my research is to enable the creation of high performace and efficient systems for large-scale data-intensive computing.
To this end, my work follows two broad approaches:
  1. Radically rethinking datacenter architecture

  2. Build robust distributed systems and protocols


Publications

Misc.
Intro | Publications | Misc.