고속 엔터프라이즈 네트워크에서 성능 저하 특성 규명

On the Performance Degradation Characteristics of High-Speed Enterprise Network

  • 주홍택 (계명대학교 컴퓨터공학과) ;
  • 홍성철 (포항공과대학교 컴퓨터공학과) ;
  • 홍원기 (포항공과대학교 컴퓨터공학과)
  • 발행 : 2009.11.30

초록

기업이나 대학 등 대규모 기관들은 고속 네트워크를 구축하여 운영하고 있다. 이와 같은 고속 엔터프라이즈 네트워크 내에서 네트워크 이용률(utilization)이 저조함에도 불구하고 종단간 성능은 네트워크 속도에 비하여 상당히 저조하다. 이러한 원인은 특정 장비에서 아주 작은 시간에 트래픽이 폭주하는 미세폭주(micro-congestion)가 발생하기 때문이다. 미세폭주는 패킷손실(Packet Loss)나 지연(Delay), 패킷역전(Packet Reordering)을 발생시키고 이것이 성능 저하의 원인이 된다. 본 논문에서는 미세폭주를 발생하는 시점을 검출하는 방법과 미세폭주를 검출한 시점에 트래픽을 수집하여 패킷손실, 지연, 패킷역전과 트래픽의 특성과의 상관관계 분석하여 성능저하를 발생시키는 미세폭주의 특성을 규명한다.

ISPs and Enterprises are equipping their networks with sufficiently high speed facilities and provide large bandwidths members. However the high speed enterprise network does not have satisfying end-to-end network performance within the network in spite of under utilization. The root cause of this performance degradation is a micro-congestion, which is a short-live event of traffic congestion. A micro-congestion causes packet loss, delay and packet reodering, and finally results in end-to-end network performance degradation. In this paper, we propose a micro-congestion detection method and find out the characteristics of performance degradation by analyzing traffic archives which is collected from a network link when a micro-congestion occurs.

키워드

참고문헌

  1. Konstantina Papagiannaki, Darryl Veitch and Nicolas Hohn, 'Origins of Microcongestion in an Access Router,' Passive & Active Measurement Workshop, Antibes, France, April, 2004
  2. Konstantina Papagiannaki, Rene Cruz and Christophe Diot, 'Network Performance Monitoring at Small Time Scales,' Internet Measurement Conference, Miami, Florida, USA, October, 2003
  3. Nicolas Hohn, Dattyl Veitch, Konstantina Papagiannaki and Christophe Diot, 'Bridging Router Performance and Queuing Theory,' ACM SIGMETRICS, New York, USA, 2004 https://doi.org/10.1145/1012888.1005728
  4. Nicolas Hohn, Konstantina Papagiannaki and Darryl Veitch, 'Capturing router congestion and delay,' IEEE/ACM Transactions on Networking, Volume 17, Issue 3, June, 2009, pp.789-802 https://doi.org/10.1109/TNET.2008.927258
  5. Klaus Mochalski, Jorg Micheel and Stephen Donnelly, 'Packet Delay and Loss at the Auckland Internet Access Path,' Passive and Active Measurement Workshop, Fort Collins, Colorado, USA, March, 2002
  6. Seung-Hwa Chung, Deepali Agrawal, Myung-Sup Kim, James W. Hong and Kihong Park, 'Analysis of Bursty Packet Loss Characteristics on Underutilized Links Using SNMP,' E2EMON, San Diego, California, USA, October, 2004
  7. Seung-Hwa Chung, Young J. Won, Deepali Agrawal, Seong-Cheol Hong, James W. Hong, Hong-Taek Ju and Kihong Park, 'Detection and Analysis of Packet Loss on Underutilized Enterprise Networks,' E2EMON, Nice, France, May, 2005 https://doi.org/10.1109/E2EMON.2005.1564477
  8. Zhi-Li Zhang, Vinay J. Ribeiro, Sue Moon and Christophe Diot, 'Small-Time Scaling Behaviors of Internet Backbone Traffic: An Empirical Study,' INFOCOM, San Francisco, USA, March, 2003 https://doi.org/10.1109/INFCOM.2003.1209205
  9. Wenyo Jiang and Henning Schulzrinne, 'Modeling of Packet Loss and Delay and Their Effect on Real-Time Multimedia Service Quality,' ACM NOSSDAV, Chapel Hill, North Carolina, June, 2000
  10. Velibor Markovski, Fei Xue and Ljiljana Trajkovic, 'Simulation and Analysis of Packet Loss using User Datagram Protocol,' The Journal of Supercomputing, Kluwer, Vol.20, No.2, pp.175-196, September, 2001 https://doi.org/10.1023/A:1011134903609
  11. Endace, DAG 4.3GE card, User Manual, EDM01.02-01
  12. Cisco, 'MIB compilers and Loading MIBs,' Cisco Technical NOtes, http://www.cisco.com/en/US/tech/tk648/tk362/technologies_tech_note09186a00800b4cee.shtml
  13. Cisco, 'Input Queue Overflow on an Interface,' Cisco Technical Notes, http://www.cisco.com/en/US/products/hw/modules/ps2643/products_tech_note09186a0080094a8c.shtml