Associate Professor
Computer Science and Engineering
EECS Department
University of Michigan
harshavm@umich.edu
Brief Bio
Update: After eight wonderful years at Michigan, I have recently moved to USC where I am now a faculty member in Computer Science.
I had been a faculty member at Michigan since 2014. Prior to arriving at Michigan, I served on the faculty at UC Riverside for four years.
My research has received the IRTF’s Applied Networking Research Prize twice and led to award papers at OSDI, NSDI, IMC, and IEEE CNS. I have also received multiple Google Faculty Research awards, a NetApp Faculty Fellowship, a Facebook Faculty Award, and an NSF CAREER award.
I am a recipient of Michigan College of Engineering’s Neil Van Eenam Memorial Undergraduate Teaching Award.
Research Overview
My research is broadly in the area of distributed systems. In particular, I view most of my work as enabling systems support for facilitating the exchange of information within human society. Over the last decade, my group has focused on this topic primarily in two settings:
- Enabling cost-effective service deployments which span the globe
- Near-Optimal Latency Versus Cost Tradeoffs in Geo-Distributed Storage, NSDI 2020
- Engineering Egress with Edge Fabric: Steering Oceans of Content to the World, SIGCOMM 2017
- CosTLO: Cost-Effective Redundancy for Lower Latency Variance on Cloud Storage Services, NSDI 2015
- SPANStore: Cost-Effective Geo-Replicated Storage Spanning Multiple Cloud Services, SOSP 2013
- Improving the web’s performance and availability
- Horcrux: Automatic JavaScript Parallelism for Resource-Efficient Web Computation, OSDI 2021
- Network Error Logging: Client-side measurement of end-to-end web service reliability, NSDI 2020
- Vroom: Accelerating the Mobile Web with Server-Aided Dependency Resolution, SIGCOMM 2017
- Klotski: Reprioritizing Web Content to Improve User Experience on Mobile Devices, NSDI 2015
More recently, I have been investigating new forms of facilitating and benefiting from information exchange in the following settings:
- Sharing human insights inferred from users’ actions (Federated learning, OSDI 2021)
- Revisiting use of cloud and edge services in manufacturing (Smart manufacturing, HotMobile 2020)
- Preserving information on the web across generations
Research Impact
Beyond papers, I strive to ensure that my work has broad real-world impact. To date, some of the ways in which my research has proved useful to users, service providers, and researchers are as follows:
- Network Error Logging (NSDI 2020) is a proposed W3C standard, which is implemented in the Google Chrome and Microsoft Edge web browsers
- Vroom (SIGCOMM 2017) is in use on the mobile website of 1-800-Flowers
- Edge Fabric (SIGCOMM 2017) enabled Facebook to perform capacity- and performance-aware routing of content to its users
- MyPageKeeper (USENIX Security 2012), an application for detecting social spam and malware on Facebook, was used by over 20,000 users, and its technology was licensed by Cloudflare
- WhyHigh (IMC 2009) helped reduce latencies to Google from hundreds to tens of milliseconds for millions of users
- iPlane (OSDI 2006, NSDI 2009) was a system I operated from 2006 to 2016, and its data was used in over a hundred research projects