Abstract
dc:description.abstractThis dissertation focuses on the operating system level requirements for advancing software elasticity (i.e. rapid scalability in fine-grain increments) in cloud applications. We assert that the highly-elastic behaviors exhibited by cloud applications today are being inhibited by the complex and inflexible layers of legacy OS and middleware software on which the applications are deployed. To this, we investigate a new system architecture designed around the first-order notion of software elasticity. The research contributions of this dissertation are made in the context of the Scalable Elastic System Architecture (SESA). Specifically, we describe the distributed runtime layer, EbbRT, which acts as our platform for designing and deploying highly-elastic applications. We demonstrate the impact of the SESA model via EbbRT across three application case studies: a high-performance in-memory cache, a CPU-bound image processing algorithm, and a serverless execution platform for a shared cloud environment.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Computer Science
- Grantor
- Boston University
- Year dc:date.issued
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Cadden, James
- Advisors dc:contributor.advisor
-
- Appavoo, Jonathan
- Matta, Abraham
- ORCID dc:identifier.orcid
Subjects
dc:subject × 4Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/2144/53506