全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于LRU淘汰机制的自适应大流检测算法

, PP. 1159-1164

Keywords: 计算机应用,网络流量测量,大流检测,自适应算法,最近最久未用

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对现有大流检测算法自适应能力差和难以满足工程应用需求的问题,提出一种新的基于“最近最久未用”淘汰机制的自适应大流检测算法。该算法设置流归并和LRU两级缓存,数据分组到达时,首先进入流归并缓存,按照“流关键字”通过哈希算法实现数据分组到流的匹配,并对流大小进行估计;同时根据上一时刻被LRU淘汰流的大小实时调整LRU缓存之前的过滤门限;然后比较流估计值和门限大小,估计值大于门限的流所含数据分组进入LRU缓存进一步筛选,否则丢弃。为保证实时性和过滤效果,分析并提出了门限时长的设置方法。理论推导和实验结果表明:该算法既保证了准确性又提高了自适应性,更适合工程应用。

References

[1]  Estan C, Varghese G, Fisk M. Bitmap algorithms for counting active flows on high speed links[C]∥Proceedings of the 3rd ACM SIGCOMM Conference on Internet Measurement, ACM, 2003: 153-166.
[2]  张震, 汪斌强, 陈庶樵, 等. 基于多维计数型布鲁姆过滤器的大流检测机制[J]. 电子与信息学报, 2010, 32(7):1608-1613. Zhang Zhen, Wang Bin-qiang, Chen Shu-qiao, et al. A mechanism of identifying heavy hitters based on multi-dimensional counting bloom filter[J]. Journal of Electronics & Information Technology, 2010, 32(7):1608-1613.
[3]  王宏, 龚正虎. Hits和Holds:识别大象流的两种算法[J]. 软件学报, 2010, 21(6):1391-1403. Wang Hong, Gong Zheng-hu. Hits and holds: two algorithms for identifying the elephant flows[J]. Journal of Software, 2010, 21(6):1391-1403.
[4]  Zhang Y, Breslau L, Paxson V, et al. On the characteristics and origins of internet flow rates[C]∥ACM SIGCOMM, Pittsburgh, PA, USA, 2002:309-322.
[5]  Feldmann A, Greenberg A, Lund C, et al. Deriving traffic demands for operational IP networks: methodology and experience[J]. IEEE/ACM Transactions on Networking, 2001, 9(3):265-280.
[6]  Zadnik M, Canini M, Moore A W, et al. Tracking elephant flows in internet backbone traffic with an fpga-based cache[C]∥International Conference on Field Programmable Logic and Applications, IEEE, 2009: 640-644.
[7]  Rong Q, Zhang G X, Xie G G. Mnemonic lossy counting: an efficient and accurate heavy-hitters identification algorithm[C]∥The 29th International Performance Computing and Communications Conference, IEEE, 2010:233-255.
[8]  Cristian E, George V. New direction in traffic measurement and accounting[J]. SIGOMM Computer Communication Review, 2002, 32(4):323-336.
[9]  Hong W, Zheng H G, Qing G, et al. Detection network anomalies based on packet and flow analysis[C]∥Seventh International Conference on Networking, IEEE, 2008: 497-502.
[10]  谢冬青, 周再红, 骆嘉伟. 基于LRU和SCBF的大象流提取及其在DDoS防御中的应用[J]. 计算机研究与发展, 2011(8):1517-1523. Xie Dong-qing, Zhou Zai-hong, Luo Jia-wei. An algorithm based on LRU and SCBF for elephant flows identification and its application in DDoS defense[J]. Journal of Computer Research and Development, 2011(8):1517-1523.
[11]  Cisco Netflow.[DB/OL].[2006-07-19].http://www.cisco.com/en/US/products/ps6601/products_ios_protocol_group_home.html.
[12]  Kim I, Reddy A L N. Identifying long-term high-bandwidth flows at a router[C]∥Proceedings of the 8th International Conference on High Performance, Berlin, Springer, 2001: 361-371.
[13]  裴育杰, 王洪波, 程时端. 基于两级LRU机制的大流检测算法[J].电子学报, 2009, 37(4):684-691. Pei Yu-jie, Wang Hong-bo, Cheng Shi-duan. A dual-LRU based algorithm for identifying and measuring large flows[J]. Acta Electronic Sinica, 2009, 37(4): 684-691.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133