Supporting Temporal Views in a Management Information Base [Vol. 7, No. 2, 1999]

Supporting Temporal Views in a Management Information Base

Vassilis J. Tsotras
Department of Computer Science,
University of California,
Riverside, CA 92521
Email: tsotras_AT_cs.ucr.edu

Vidyadhar Phalke
Network Programs Inc.,
255 Old New Brunswick Rd.,
Piscataway, NJ 08854

Anil Kumar
Department of Electrical Engineering,
Polytechnic University,
Brooklyn, NY 11201

B. Gopinath
Network Programs Inc.,
255 Old New Brunswick Rd.,
Piscataway, NJ 08854

Abstract
In many network management applications, like post-mortem fault analysis or performance trends profiling, it is advantageous to have the ability to view the state of the network as it was at some time in the past. To support such Temporal Views an efficient data organization, or access method, is needed for storing and updating network related data (as the network evolves over time) and for retrieving requested past network states. For applications where the network manager is not interested in the full (and maybe too large) snapshot of a past network state it is useful if partial state snapshots can be extracted quickly. It is thus of particular interest to construct an access method that can efficiently support Partial Temporal Views. Efficiency implies that a requested partial temporal view should be constructed directly, without first computing the elaborate full temporal view. In this paper we present a new access method (called the Neighbor History Index) for this problem. One of the advantages of this method is that the update processing is independent of the evolution size (the total number of changes in the evolution). In addition, our method uses a small disk space overhead. We then present a general framework for organizing time-evolving network data. Our framework distinguishes between flat and hierarchical evolutions and subsequently between flat and hierarchical temporal views. We also provide a way to efficiently construct temporal views on hierarchical evolutions. This paper shows that supporting temporal views on flat or hierarchical evolutions is not expensive: our solutions use small space overhead, have small updating and compute temporal views fast.

Keywords: Network Management; Temporal Databases; Historical Queries; Data Organization.

JNSM: Vol. 7, No. 2, 1999 Supporting Temporal Views in a Management Information Base [Vol. 7, No. 2, 1999]

NOTE: only abstract of paper available on-line

Back to JNSM main page