The dissertation of Alexander Fuerst, who recently earned a Luddy School of Informatics, Computing, and Engineering Intelligent Systems Engineering Ph.D., placed second in the prestigious SPEC Kaivalya Dixit Distinguished Dissertation Award competition.
The award recognizes outstanding doctoral dissertations in the field of computer benchmarking, performance evaluation, and experimental system analysis.
Nominated dissertations were evaluated for scientific originality, scientific significance, practical relevance, impact, and quality of the presentation.
Standard Performance Evaluation Corporation (SPEC) is the main international industry and academia group for performance engineering. It’s a platform for collaborative research efforts in quantitative system evaluation and analysis.
Fuerst said his dissertation, “Serverless Control Planes for Orchestration of Cloud Resources,” addressed serverless computing performance and resource management challenges by designing systems and tailoring algorithms to handle those workloads.
He said his research orchestrates resource allocation and scheduling at various levels to account for heterogenous characteristics. The highlight of it is Ilúvatar, a redesigned and disaggregated serverless control plane to improve research capabilities.
“Alex’s thesis work identifies and addresses many of the fundamental and practical performance challenges in large-scale Functions as a Service (FaaS),” said Prateek Sharma, associate professor of Intelligent Systems Engineering, director of graduate studies for Intelligent Systems Engineering and Fuerst’s Ph.D. advisor. “FaaS is a new paradigm for cloud-native applications and is being used by applications such as web services, machine learning, data analytics, and high-performance computing.”
Dissertation award officials said they normally recognize just one dissertation, but the quality of the submissions caused them to honor two others as runner-up, including Fuerst’s.
“Receiving runner-up for the SPEC dissertation is a significant honor for Dr. Fuerst,” said Beth Plale, Intelligent Systems Engineering chair and Michael A and Laurie Burns McRobbie Bicentennial Professor of Computer Engineering. “His dissertation topic is timely – an energy-sensitive control plane for executing functions on cloud resources. Our department is very proud of him.”
Added Fuerst: “I never expected my dissertation to get traction in such a far-flung area as SPEC. Hearing that they thought my work was such high quality makes me proud of what went into it.”
Sharma said the thesis’s “piece de resistance” was the Illuvatar software system.
“It’s designed from the ground-up to serve as an extendible platform for performance research,” Sharma said.
“Through its algorithmic and design and implementation advances, this dissertation improves function latency by more than 100 times, and provides the firm conceptual and infrastructural foundations for high-performance FaaS research and deployments.”
Fuerst, who has been offered a job with the Department of Defense, continues to conduct research with Sharma.
Fuerst’s research interests encompass the entire computing system stack. That includes various segments of systems research, especially Serverless, Cloud, and Virtualization.
The SPEC Kaivalya Dixit Distinguished Dissertation Award was established in 2011 to recognize outstanding dissertations within the scope of the SPEC Research Group. This spans the design of metrics for system evaluation and the development of methodologies, techniques, and tools for measurement, load testing, profiling, workload characterization, dependability, and efficiency evaluation of computing systems.