University of Illinois Urbana-Champaign

Nimblock 2.0: Serverless computing with multi-tenant FPGAS

Mandava, Meghna

This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.

Permalink

https://hdl.handle.net/2142/120445

Description

Title
Nimblock 2.0: Serverless computing with multi-tenant FPGAS
Author(s)
Mandava, Meghna
Issue Date
2023-05-01
Director of Research (if dissertation) or Advisor (if thesis)
Chen, Deming
Department of Study
Computer Science
Discipline
Computer Science
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
  • FPGA
  • Serverless Computing
  • Reconfigurable Computing
  • Heterogeneous Cloud
Abstract
Serverless computing has gained significant popularity in recent years, especially in cloud computing environments. Although Field-Programmable Gate Arrays (FPGAs) have become increasingly common in cloud computing environments, there has been limited research into how FPGAs can be integrated into serverless computing frameworks to further improve performance and cost-efficiency. We propose Nimblock 2.0 as a means of integrating FPGAs into serverless platforms, enabling developers to leverage the speed and versatility of FPGAs to build and execute high-performance, customizable applications easily. Our evaluation of the Nimblock 2.0 heterogeneous serverless computing model shows that running FPGA workloads in a serverless environment results in an average overhead of only 13%. Allocating requests to both CPU and FPGA compute resources can provide up to a 12% performance improvement. The ability to allocate requests to different resource types can provide significant performance improvements, while considering workload characteristics and priority can further optimize performance. Nimblock 2.0 can enable the efficient use of FPGAs in serverless computing, enabling support for frameworks such as Library-as-a-Service.
Graduation Semester
2023-05
Type of Resource
Thesis
Copyright and License Information
Copyright 2023 Meghna Mandava

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois