Estimating TCP Packet Loss Ratio from Sampled ACK Packets
スポンサーリンク
概要
- 論文の詳細を見る
The advent of various quality-sensitive applications has greatly changed the requirements for IP network management and made the monitoring of individual traffic flows more important. Since the processing costs of per-flow quality monitoring are high, especially in high-speed backbone links, packet sampling techniques have been attracting considerable attention. Existing sampling techniques, such as those used in Sampled NetFlow and sFlow, however, focus on the monitoring of traffic volume, and there has been little discussion of the monitoring of such quality indexes as packet loss ratio. In this paper we propose a method for estimating, from sampled packets, packet loss ratios in individual TCP sessions. It detects packet loss events by monitoring duplicate ACK events raised by each TCP receiver. Because sampling reveals only a portion of the actual packet loss, the actual packet loss ratio is estimated statistically. Simulation results show that the proposed method can estimate the TCP packet loss ratio accurately from a 10% sampling of packets.
- (社)電子情報通信学会の論文
- 2008-02-01
著者
-
SHIMONISHI Hideyuki
System Platforms Laboratory, NEC Corporation
-
Murase Tutomu
System Platforms Laboratory Nec Corporation
-
Yamasaki Yasuhiro
System Platforms Research Laboratories Nec Corporation
-
Shimonishi Hideyuki
System Platforms Research Laboratories Nec Corporation
-
Shimonishi Hideyuki
System Platforms Laboratory Nec Corporation
関連論文
- TCP-Westwood Low-Priority for Overlay QoS Mechanism(Networking Technologies for Overlay Networks)
- Hierarchically Aggregated Fair Queueing (HAFQ) for Per-Flow Fair Bandwidth Allocation(Switching for Communications)
- SB-10-1 Congestion Prevention Buffer Management in TCP Proxy(SB-10. Latest Trends on Broadband Contents Delivery Technologies)
- Estimating TCP Packet Loss Ratio from Sampled ACK Packets
- BS-3-22 A Novel Higher Wireless Connection-Capacity Route Selection Algorithm for Automobiles(BS-3. Management and Control Technologies for Innovative Networks)
- Architecture, Implementation, and Experiments of Programmable Network Using OpenFlow
- Fully Distributed Self-Organization of Shortest Spanning Tree and Optimal Sink Node Position for Large-Scale Wireless Sensor Network
- Novel Access-Point Selection for User QoS and System Optimization Based on User Cooperative Moving
- Route Instruction Mechanism for Mobile Users Leveraging Distributed Wireless Resources