Research
PhD Research
VIVA
An end-to-end system for interactive video analytics.
[Publication (vision)] [Publication (full)] [Preprint]
Llama
A heterogeneous and serverless framework for auto-tuning video pipelines.
Faa$T
A transparent auto-scaling distributed cache for serverless applications. Faa$T's design is motivated by a comprehensive study of data access patterns in Azure Functions. This project was in collaboration with Microsoft Research. A version of Faa$T is available in production Azure Functions.
[Publication] [Github Repository 1] [Github Repository 2]
INFaaS
An inference-as-a-service system that abstracts resource management and model selection.
[Publication (full)] [Publication (system vision)] [Publication (scheduler vision)] [Website]
gg
A highly-scalable, function-as-a-service orchestration management framework for applications that are normally run on a laptop or on a cluster that can stay idle for long periods of time (e.g., video encoding and compilation).
[Publication] [Github Repository]
In a previous life...
Methods for Selecting the Right Storage Technology for Larger-Than-Memory Databases
with David Cheriton
Mage: Online and Interference-Aware Scheduling for Multi-Scale Heterogeneous Systems
with Christina Delimitrou
Behavioral Signal Processing
with Shrikanth Narayanan
General Purpose GPU Graph Applications
with Murali Annavaram
Propagation Characteristics of Heterogeneous Composites of Materials at 60 GHz
with John Wawrzynek
This work was done as a part of the 2014 University of California, Berkeley SUPERB-ITS REU. It was presented at the 2014 SACNAS National Conference in Los Angeles, California, where it won a best poster award.