Secure cache provision: Provable DDOS prevention for randomly partitioned services with replication

Weibo Chu; Xiaohong Guan; John C.S. Lui; Zhongmin Cai; Xiaohong Shi

doi:10.1109/ICDCSW.2013.24

Secure cache provision: Provable DDOS prevention for randomly partitioned services with replication

Weibo Chu
, Xiaohong Guan
, John C.S. Lui
, Zhongmin Cai
, Xiaohong Shi

Faculty of Electronic and Information Engineering

Xi'an Jiaotong University
Tsinghua University
Chinese University of Hong Kong
Beijing Qihoo Technology Co. Ltd.

Research output: Contribution to conference › Paper › peer-review

Abstract

In this paper we show a small but fast popularitybased front-end cache can provide provable DDOS prevention for randomly partitioned cluster services with replication. To achieve this, we first give the best strategy for an adversary to overload the system, and then prove that the cache size is lower bounded by O(n log log n/ log d), where n is the number of back-end nodes and d is the replication factor. Since log log n/ logd < 2 holds for almost all the current clusters (i.e., the number of back-end nodes n < 105 and the replication factor d &ge; 3), this result implies an O(n) lower bound on the required cache size. Our analysis and results are well validated through extensive simulations.

Original language	English
Pages	58-63
Number of pages	6
DOIs	https://doi.org/10.1109/ICDCSW.2013.24
State	Published - 2013
Event	33rd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2013 - Philadelphia, PA, United States Duration: 8 Jul 2013 → 11 Jul 2013

Conference

Conference	33rd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2013
Country/Territory	United States
City	Philadelphia, PA
Period	8/07/13 → 11/07/13

Access to Document

10.1109/ICDCSW.2013.24

Cite this

@conference{39b712bb2b7e45bfa89fa7444a350e70,

title = "Secure cache provision: Provable DDOS prevention for randomly partitioned services with replication",

abstract = "In this paper we show a small but fast popularitybased front-end cache can provide provable DDOS prevention for randomly partitioned cluster services with replication. To achieve this, we first give the best strategy for an adversary to overload the system, and then prove that the cache size is lower bounded by O(n log log n/ log d), where n is the number of back-end nodes and d is the replication factor. Since log log n/ logd < 2 holds for almost all the current clusters (i.e., the number of back-end nodes n < 105 and the replication factor d \≥ 3), this result implies an O(n) lower bound on the required cache size. Our analysis and results are well validated through extensive simulations.",

author = "Weibo Chu and Xiaohong Guan and Lui, \{John C.S.\} and Zhongmin Cai and Xiaohong Shi",

year = "2013",

doi = "10.1109/ICDCSW.2013.24",

language = "英语",

pages = "58--63",

note = "33rd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2013 ; Conference date: 08-07-2013 Through 11-07-2013",

}

Secure cache provision: Provable DDOS prevention for randomly partitioned services with replication. / Chu, Weibo; Guan, Xiaohong; Lui, John C.S. et al.
2013. 58-63 Paper presented at 33rd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2013, Philadelphia, PA, United States.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Secure cache provision

T2 - 33rd IEEE International Conference on Distributed Computing Systems Workshops, ICDCSW 2013

AU - Chu, Weibo

AU - Guan, Xiaohong

AU - Lui, John C.S.

AU - Cai, Zhongmin

AU - Shi, Xiaohong

PY - 2013

Y1 - 2013

N2 - In this paper we show a small but fast popularitybased front-end cache can provide provable DDOS prevention for randomly partitioned cluster services with replication. To achieve this, we first give the best strategy for an adversary to overload the system, and then prove that the cache size is lower bounded by O(n log log n/ log d), where n is the number of back-end nodes and d is the replication factor. Since log log n/ logd < 2 holds for almost all the current clusters (i.e., the number of back-end nodes n < 105 and the replication factor d ≥ 3), this result implies an O(n) lower bound on the required cache size. Our analysis and results are well validated through extensive simulations.

AB - In this paper we show a small but fast popularitybased front-end cache can provide provable DDOS prevention for randomly partitioned cluster services with replication. To achieve this, we first give the best strategy for an adversary to overload the system, and then prove that the cache size is lower bounded by O(n log log n/ log d), where n is the number of back-end nodes and d is the replication factor. Since log log n/ logd < 2 holds for almost all the current clusters (i.e., the number of back-end nodes n < 105 and the replication factor d ≥ 3), this result implies an O(n) lower bound on the required cache size. Our analysis and results are well validated through extensive simulations.

UR - https://www.scopus.com/pages/publications/84893303358

U2 - 10.1109/ICDCSW.2013.24

DO - 10.1109/ICDCSW.2013.24

M3 - 论文

AN - SCOPUS:84893303358

SP - 58

EP - 63

Y2 - 8 July 2013 through 11 July 2013

ER -

Secure cache provision: Provable DDOS prevention for randomly partitioned services with replication

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this