CSE alum Akshitha Sriraman receives ProQuest Distinguished Dissertation Award
CSE alum Akshitha Sriraman has been selected as the recipient of the ProQuest Distinguished Dissertation award from the Rackham Graduate School for her dissertation, “Enabling Hyperscale Web Services.” The award recognizes the most exceptional scholarly work produced by a doctoral student in any field of study at the University of Michigan.
Sriraman’s dissertation explores how to bridge computer architecture and software systems to tackle new challenges posed by massive web services, such as social media, online messaging, web search, video streaming, gaming, and online banking. These services serve billions of users and require data centers that scale to hundreds of thousands of servers. Services and data centers of this size are often called hyperscale.
In her dissertation, Sriraman presented technologies to enable these growing services to handle the massive user load while still offering good performance, cost, and energy efficiency. She tackled two key challenges facing hyperscale web services – the rapid, sudden growth of data and user base, and the decline in hardware performance scaling.
Sriraman demonstrated that systems researchers will need to abandon the traditional approach of building each layer of the systems stack separately in favor of a new synergy between hardware and software systems. According to her dissertation, this synergy will involve software that is aware of new hardware constraints and hardware that is architected to efficiently support new hyperscale software requirements. She developed several tools to address both prongs of this solution, including benchmark suites, redesigned threading models, characterizations and demonstrations of existing hardware limitations and capabilities, a systematic understanding of hardware customization opportunities, and an analytical alternative to ad hoc hardware customization.
Sriraman is now an Assistant Professor at Carnegie Mellon University in the Electrical and Computer Engineering Department. There, she continues her work on designing software that is aware of new hardware constraints/possibilities and architecting hardware that efficiently supports hyperscale software requirements. She was advised by Prof. Thomas Wenisch.