Line speed heavy hitters on sliding windows

Ran Ben Basat; Gil Einziger; Erez Waisbard

doi:10.1109/INFCOMW.2018.8406850

Line speed heavy hitters on sliding windows

Ran Ben Basat, Gil Einziger, Erez Waisbard

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Sliding window measurements are required by network functionalities such as load balancing, traffic engineering, and intrusion detection. Previous works suggested finding the most frequent flows, also known as heavy hitters but the suggested algorithms are too slow for modern lines. This work introduces a novel randomized algorithm named HHAWAII that accelerates update time by performing lightweight updates for most packets. HHAWAII is evaluated on three real packet traces, it achieves a similar accuracy to previous algorithms and as much as 14x higher update throughput.

Original language	English
Title of host publication	INFOCOM 2018 - IEEE Conference on Computer Communications Workshops
Publisher	Institute of Electrical and Electronics Engineers
Pages	1-2
Number of pages	2
ISBN (Electronic)	9781538659793
DOIs	https://doi.org/10.1109/INFCOMW.2018.8406850
State	Published - 6 Jul 2018
Externally published	Yes
Event	2018 IEEE Conference on Computer Communications Workshops, INFOCOM 2018 - Honolulu, United States Duration: 15 Apr 2018 → 19 Apr 2018

Publication series

Name	INFOCOM 2018 - IEEE Conference on Computer Communications Workshops

Conference

Conference	2018 IEEE Conference on Computer Communications Workshops, INFOCOM 2018
Country/Territory	United States
City	Honolulu
Period	15/04/18 → 19/04/18

ASJC Scopus subject areas

Control and Optimization
Artificial Intelligence
Computer Networks and Communications

Access to Document

10.1109/INFCOMW.2018.8406850

Cite this

@inproceedings{29b97f42642241febd3bf392bb7d6956,

title = "Line speed heavy hitters on sliding windows",

abstract = "Sliding window measurements are required by network functionalities such as load balancing, traffic engineering, and intrusion detection. Previous works suggested finding the most frequent flows, also known as heavy hitters but the suggested algorithms are too slow for modern lines. This work introduces a novel randomized algorithm named HHAWAII that accelerates update time by performing lightweight updates for most packets. HHAWAII is evaluated on three real packet traces, it achieves a similar accuracy to previous algorithms and as much as 14x higher update throughput.",

author = "{Ben Basat}, Ran and Gil Einziger and Erez Waisbard",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Conference on Computer Communications Workshops, INFOCOM 2018 ; Conference date: 15-04-2018 Through 19-04-2018",

year = "2018",

month = jul,

day = "6",

doi = "10.1109/INFCOMW.2018.8406850",

language = "English",

series = "INFOCOM 2018 - IEEE Conference on Computer Communications Workshops",

publisher = "Institute of Electrical and Electronics Engineers",

pages = "1--2",

booktitle = "INFOCOM 2018 - IEEE Conference on Computer Communications Workshops",

address = "United States",

}

Ben Basat, R, Einziger, G & Waisbard, E 2018, Line speed heavy hitters on sliding windows. in INFOCOM 2018 - IEEE Conference on Computer Communications Workshops. INFOCOM 2018 - IEEE Conference on Computer Communications Workshops, Institute of Electrical and Electronics Engineers, pp. 1-2, 2018 IEEE Conference on Computer Communications Workshops, INFOCOM 2018, Honolulu, United States, 15/04/18. https://doi.org/10.1109/INFCOMW.2018.8406850

Line speed heavy hitters on sliding windows. / Ben Basat, Ran; Einziger, Gil; Waisbard, Erez.
INFOCOM 2018 - IEEE Conference on Computer Communications Workshops. Institute of Electrical and Electronics Engineers, 2018. p. 1-2 (INFOCOM 2018 - IEEE Conference on Computer Communications Workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Line speed heavy hitters on sliding windows

AU - Ben Basat, Ran

AU - Einziger, Gil

AU - Waisbard, Erez

PY - 2018/7/6

Y1 - 2018/7/6

N2 - Sliding window measurements are required by network functionalities such as load balancing, traffic engineering, and intrusion detection. Previous works suggested finding the most frequent flows, also known as heavy hitters but the suggested algorithms are too slow for modern lines. This work introduces a novel randomized algorithm named HHAWAII that accelerates update time by performing lightweight updates for most packets. HHAWAII is evaluated on three real packet traces, it achieves a similar accuracy to previous algorithms and as much as 14x higher update throughput.

AB - Sliding window measurements are required by network functionalities such as load balancing, traffic engineering, and intrusion detection. Previous works suggested finding the most frequent flows, also known as heavy hitters but the suggested algorithms are too slow for modern lines. This work introduces a novel randomized algorithm named HHAWAII that accelerates update time by performing lightweight updates for most packets. HHAWAII is evaluated on three real packet traces, it achieves a similar accuracy to previous algorithms and as much as 14x higher update throughput.

UR - http://www.scopus.com/inward/record.url?scp=85050693971&partnerID=8YFLogxK

U2 - 10.1109/INFCOMW.2018.8406850

DO - 10.1109/INFCOMW.2018.8406850

M3 - Conference contribution

AN - SCOPUS:85050693971

T3 - INFOCOM 2018 - IEEE Conference on Computer Communications Workshops

SP - 1

EP - 2

BT - INFOCOM 2018 - IEEE Conference on Computer Communications Workshops

PB - Institute of Electrical and Electronics Engineers

T2 - 2018 IEEE Conference on Computer Communications Workshops, INFOCOM 2018

Y2 - 15 April 2018 through 19 April 2018

ER -

Line speed heavy hitters on sliding windows

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this