Characteristics of Temporal and Spatial Locality of Internet Access Patterns

Authors:
Keisuke Ishibashi;Masaki Aida;Makoto Imase
Affiliations:
-;-;-
Venue:
Revised Papers from the NETWORKING 2002 Workshops on Web Engineering and Peer-to-Peer Computing
Year:
2002

Citing 3
Cited 0

The Stack Growth Function: Cache Line Reference Models

IEEE Transactions on Computers
Characterizing reference locality in the WWW

DIS '96 Proceedings of the fourth international conference on on Parallel and distributed information systems
Speculative Data Dissemination and Service to Reduce Server Load, Network Traffic and Service Time in Distributed Information Systems

ICDE '96 Proceedings of the Twelfth International Conference on Data Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

The locality of access patterns is a significant characteristic to be considered in analyzing Internet access behaviors. Previous research focused on temporal locality, which implies a high probability of the same IP address reappearing. In this paper, in addition to temporal locality we analyze spatial locality behavior, which implies a high probability of neighboring IP addresses appearing. Using actual Internet traces, we have analyzed the relationship between the number of accesses and the number of distinct elements appearing for both full addresses and address prefixes. We found that the number of distinct full addresses appearing grows much faster than the number of address prefixes, and they are related by a power law. Also, we compose a stochastic model that generates an address sequence with hierarchical structure consisting of full address and address prefix. We verify that an address sequence generated by the model shows both spatial and temporal locality behaviors similar to those of actual data.