Scalable URL matching with small memory footprint

Anat Bremler-Barr, David Hay, Daniel Krauthgamer, Shimrit Tzur-David

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

URL matching lies at the core of many networking applications and Information Centric Networking architectures. For example, URL matching is extensively used by Layer 7 switches, ICN/NDN routers, load balancers, and security devices. Modern URL matching is done by maintaining a rich database that consists of tens of millions of URL which are classified to dozens of categories (or egress ports). In real-time, any input URL has to be searched in this database to find the corresponding category. In this paper, we introduce a generic framework for accurate URL matching (namely, no false positives or miscategorization) that aims to reduce the overall memory footprint, while still having low matching latency. We introduce a dictionary-based compression method that compresses the database by 60%, while having only a slight overhead in time. Our framework is very flexible and it allows hot-updates, cloud-based deployments, and can deal with strings that are not URLs.

Original languageAmerican English
Title of host publication2016 IFIP Networking Conference (IFIP Networking) and Workshops, IFIP Networking 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages467-475
Number of pages9
ISBN (Electronic)9783901882838
DOIs
StatePublished - 21 Jun 2016
Event2016 IFIP Networking Conference (IFIP Networking) and Workshops, IFIP Networking 2016 - Vienna, Austria
Duration: 17 May 201619 May 2016

Publication series

Name2016 IFIP Networking Conference (IFIP Networking) and Workshops, IFIP Networking 2016

Conference

Conference2016 IFIP Networking Conference (IFIP Networking) and Workshops, IFIP Networking 2016
Country/TerritoryAustria
CityVienna
Period17/05/1619/05/16

Bibliographical note

Publisher Copyright:
© 2016 IFIP.

Fingerprint

Dive into the research topics of 'Scalable URL matching with small memory footprint'. Together they form a unique fingerprint.

Cite this