CAREER: Algorithms for Organizing and Scheduling Distributed Resources

Northeastern University, Boston MA

Investigators

Abstract

CAREER: Algorithms for Organizing and Scheduling Distributed Resources This project encompasses several basic algorithmic problems concerning the organization and scheduling of shared resources in distributed environments. Over the course of this project, problems applying to a number of different environments will be studied, including wide-area networks, asymmetric communication channels, broadcast media, and wireless networks. An important component of this project is the study of replication, which is an effective paradigm in the design of high-performance scalable information systems. Two major goals of this project are (i) the design of a provably efficient algorithm for coordinated placement of replicated objects, and (ii) the design of a distributed service for directing access requests to appropriately chosen servers. Once an access request is assigned to a server, the end-to-end performance observed by the request critically depends on the scheduling policy of the server and the allocation of network resources for the delivery of data from the server to the client. This project also explores competitive online algorithms for server scheduling in unicast and broadcast environments, and the approximability of certain NP-hard optimization problems arising in unicast and multicast routing. The final research component of this project focuses on two important aspects related to resource management in wireless networks: contention and mobility. Contention resolution protocols for uplink scheduling in cellular networks and the design of basic control structures in mobile ad-hoc networks will be studied. The integrated educational goals of this project include (i) experimental projects in cooperative caching and distributed directory services that will provide research experience for undergraduate students, (ii) the development of a new undergraduate course on the organization, representation, maintenance, and retrieval of data in diverse computing models, and (iii) the development of an advanced graduate course on algorithmic issues in mobile computing.