Delay, stability, and resource tradeoffs in large distributed service systems

This thesis addresses fundamental tradeoffs in the design of dispatching policies in large-scale distributed service systems, motivated by examples such as cloud computing facilities and multi-core processors. A canonical framework for modeling such systems is provided by a parallel queueing model with n servers, where service requests arrive stochastically over time as a single stream of jobs of rate proportional to n, and where a central controller is responsible for all decisions. The central controller makes decisions based on limited information about the state of the queues, which is conveyed through messages from servers to the dispatcher, and stored in a limited local memory. Our objective is to understand the best possible performance of such systems (in terms of stability region and delay) and to propose optimal policies, with emphasis on the asymptotic regime when both the number of servers and the arrival rate are large.

Name thesis:degree_name

Doctoral

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2019

Author dc:creator

• Zubeldía Suárez, Martín.

Advisor dc:contributor.advisor

• David Gamarnik and John N. Tsitsiklis.

• Electrical Engineering and Computer Science.

Statement dc:rights

• MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/124124

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/124124