Skip to main content

COVID-19 Notice

Anyone attending an indoor, in-person event must comply with the chancellor’s order on wearing masks while indoors in campus buildings or facilities, regardless of vaccination status.

Enabling Hyperscale Web Services

Akshitha Sriraraman (University of Michigan)

Event Details

Date
Friday, March 6, 2020
Time
3:30-4:30 p.m.
Location
Description

Modern hyperscale web service systems introduce trade-offs between performance and numerous features essential for cost- and energy-efficient operation of data centers (e.g., high server utilization, continuous power management, use of commodity hardware and software, etc.). In this talk, I will present two solutions to bridge the performance vs. cost and energy efficiency gap in hyperscale web services (1) a software system that auto-tunes threading models during system run-time to minimize web service tail latency (OSDI 2018) and (2) a system that exploits coarse-grained OS and hardware configuration knobs to tune limited cost-efficient commodity hardware stock keeping units, to better support their assigned service (ISCA 2019).

Bio:Akshitha Sriraman (http://akshithasriraman.eecs.umich.edu/) is a Ph.D. candidate at the University of Michigan. She is advised by Prof. Thomas F. Wenisch on her computer architecture and systems dissertation research, specifically on the topic of enabling hyperscale web services. Her work bridges computer architecture and software systems and demonstrates the importance of that bridge in improving the performance, cost, and energy efficiency of modern hyperscale data center systems. Sriraman has influenced the design of server architectures both via hardware analysis of production data center systems and her subsequent software design that uses data center hardware more efficiently; she received recognition for this work via the 2020 Facebook Distributed Systems Fellowship. Additionally, Sriraman has developed a novel software system that improves data center performance by minimizing tail latency; she was awarded the Rackham Merit Ph.D. Fellowship to help fund this work. 

Cost
Free