Resource Management in Large Systems.
Abstract
This work studies the application of decentralization to the problem of Resource Management in Large Systems. In an operational system, where a very large number of activities share limited resources, this Resource Management problem has three objectives. The first ('Initial Allocation') is to find an assignment of resources to every activity, such that all the system contraints are satisfied, and all activities are operating. The second ('New-Assignment') is to find a rationale for allocating resources to new activities. New activities are initiated frequently enough that we do not wish to re-solve the entire problem for the combined set of old and new activities. The third objective ('Periodic Review') is to find an efficient way of re-allocating resources in order to reflect the changing needs of the individual activities, as well as the changes in total resource usages. Conventionally, the resource-allocation problem has been studied for the case where, in addition to the constraints, there exists an objective to be maximized. Our emphasis, as is reflected by the title of this work, is on the feasibility aspect of the problem, that is, of taking a large system and keeping it operational. We show that this in itself is both an important problem, and has theoretically interesting consequences. In addition, our results can be useful for the solution of the general (optimization) problem.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1978
- Accession Number
- ADA064780
Entities
People
- Rajan Suri
Organizations
- Harvard University