Lossy compression of packet classifiers

Ori Rottenstreich, J. Tapolcai

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

Packet classification is a building block in many network services such as routing, filtering, intrusion detection, accounting, monitoring, load-balancing and policy enforcement. Compression has gained attention recently as a way to deal with the expected increase of classifiers size. Typically, compression schemes try to reduce a classifier size while keeping it semantically-equivalent to its original form. Inspired by the advantages of popular compression schemes (e.g. JPEG and MPEG), we study in this paper the applicability of lossy compression to create packet classifiers requiring less memory than optimal semantically-equivalent representations. Our objective is to find a limited-size classifier that can correctly classify a high portion of the traffic so that it can be implemented in commodity switches with classification modules of a given size. We develop optimal dynamic programming based algorithms for several versions of the problem and describe how a small amount of traffic that cannot be classified can be easily treated, especially in software-defined networks. We generalize our solutions for a wide range of classifiers with different similarity metrics. We evaluate their performance on real classifiers and traffic traces and show that in some cases we can reduce a classifier size by orders of magnitude while still classifying almost all traffic correctly.

Original languageEnglish
Title of host publicationANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages39-50
Number of pages12
ISBN (Print)9781467366335
DOIs
Publication statusPublished - May 18 2015
Event11th ACM/IEEE Symposium on Architectures for Networking and Communications Systems, ANCS 2015 - Oakland, United States
Duration: May 7 2015May 8 2015

Other

Other11th ACM/IEEE Symposium on Architectures for Networking and Communications Systems, ANCS 2015
CountryUnited States
CityOakland
Period5/7/155/8/15

Fingerprint

Classifiers
Intrusion detection
Dynamic programming
Resource allocation
Switches
Data storage equipment
Monitoring

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Rottenstreich, O., & Tapolcai, J. (2015). Lossy compression of packet classifiers. In ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (pp. 39-50). [7110119] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ANCS.2015.7110119

Lossy compression of packet classifiers. / Rottenstreich, Ori; Tapolcai, J.

ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems. Institute of Electrical and Electronics Engineers Inc., 2015. p. 39-50 7110119.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rottenstreich, O & Tapolcai, J 2015, Lossy compression of packet classifiers. in ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems., 7110119, Institute of Electrical and Electronics Engineers Inc., pp. 39-50, 11th ACM/IEEE Symposium on Architectures for Networking and Communications Systems, ANCS 2015, Oakland, United States, 5/7/15. https://doi.org/10.1109/ANCS.2015.7110119
Rottenstreich O, Tapolcai J. Lossy compression of packet classifiers. In ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems. Institute of Electrical and Electronics Engineers Inc. 2015. p. 39-50. 7110119 https://doi.org/10.1109/ANCS.2015.7110119
Rottenstreich, Ori ; Tapolcai, J. / Lossy compression of packet classifiers. ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 39-50
@inproceedings{fb74695193f946da8726c773cdb9be07,
title = "Lossy compression of packet classifiers",
abstract = "Packet classification is a building block in many network services such as routing, filtering, intrusion detection, accounting, monitoring, load-balancing and policy enforcement. Compression has gained attention recently as a way to deal with the expected increase of classifiers size. Typically, compression schemes try to reduce a classifier size while keeping it semantically-equivalent to its original form. Inspired by the advantages of popular compression schemes (e.g. JPEG and MPEG), we study in this paper the applicability of lossy compression to create packet classifiers requiring less memory than optimal semantically-equivalent representations. Our objective is to find a limited-size classifier that can correctly classify a high portion of the traffic so that it can be implemented in commodity switches with classification modules of a given size. We develop optimal dynamic programming based algorithms for several versions of the problem and describe how a small amount of traffic that cannot be classified can be easily treated, especially in software-defined networks. We generalize our solutions for a wide range of classifiers with different similarity metrics. We evaluate their performance on real classifiers and traffic traces and show that in some cases we can reduce a classifier size by orders of magnitude while still classifying almost all traffic correctly.",
author = "Ori Rottenstreich and J. Tapolcai",
year = "2015",
month = "5",
day = "18",
doi = "10.1109/ANCS.2015.7110119",
language = "English",
isbn = "9781467366335",
pages = "39--50",
booktitle = "ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Lossy compression of packet classifiers

AU - Rottenstreich, Ori

AU - Tapolcai, J.

PY - 2015/5/18

Y1 - 2015/5/18

N2 - Packet classification is a building block in many network services such as routing, filtering, intrusion detection, accounting, monitoring, load-balancing and policy enforcement. Compression has gained attention recently as a way to deal with the expected increase of classifiers size. Typically, compression schemes try to reduce a classifier size while keeping it semantically-equivalent to its original form. Inspired by the advantages of popular compression schemes (e.g. JPEG and MPEG), we study in this paper the applicability of lossy compression to create packet classifiers requiring less memory than optimal semantically-equivalent representations. Our objective is to find a limited-size classifier that can correctly classify a high portion of the traffic so that it can be implemented in commodity switches with classification modules of a given size. We develop optimal dynamic programming based algorithms for several versions of the problem and describe how a small amount of traffic that cannot be classified can be easily treated, especially in software-defined networks. We generalize our solutions for a wide range of classifiers with different similarity metrics. We evaluate their performance on real classifiers and traffic traces and show that in some cases we can reduce a classifier size by orders of magnitude while still classifying almost all traffic correctly.

AB - Packet classification is a building block in many network services such as routing, filtering, intrusion detection, accounting, monitoring, load-balancing and policy enforcement. Compression has gained attention recently as a way to deal with the expected increase of classifiers size. Typically, compression schemes try to reduce a classifier size while keeping it semantically-equivalent to its original form. Inspired by the advantages of popular compression schemes (e.g. JPEG and MPEG), we study in this paper the applicability of lossy compression to create packet classifiers requiring less memory than optimal semantically-equivalent representations. Our objective is to find a limited-size classifier that can correctly classify a high portion of the traffic so that it can be implemented in commodity switches with classification modules of a given size. We develop optimal dynamic programming based algorithms for several versions of the problem and describe how a small amount of traffic that cannot be classified can be easily treated, especially in software-defined networks. We generalize our solutions for a wide range of classifiers with different similarity metrics. We evaluate their performance on real classifiers and traffic traces and show that in some cases we can reduce a classifier size by orders of magnitude while still classifying almost all traffic correctly.

UR - http://www.scopus.com/inward/record.url?scp=84936074846&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936074846&partnerID=8YFLogxK

U2 - 10.1109/ANCS.2015.7110119

DO - 10.1109/ANCS.2015.7110119

M3 - Conference contribution

SN - 9781467366335

SP - 39

EP - 50

BT - ANCS 2015 - 11th 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems

PB - Institute of Electrical and Electronics Engineers Inc.

ER -