Section 1: TCP Measurements -- On the Impact of Bursting on TCP Performance -- A Study of Burstiness in TCP Flows -- On the Stationarity of TCP Bulk Data Transfers -- Section 2: Application Measurements -- Toward the Accurate Identification of Network Applications -- A Traffic Identification Method and Evaluations for a Pure P2P Application -- Analysis of Peer-to-Peer Traffic on ADSL -- Analysis of Communities of Interest in Data Networks -- Section 3: Network Inference and Problem Diagnosis -- Binary Versus Analogue Path Monitoring in IP Networks -- Exploiting the IPID Field to Infer Network Path and End-System Characteristics -- New Methods for Passive Estimation of TCP Round-Trip Times -- Detecting Duplex Mismatch on Ethernet -- Section 4: Topology Measurements -- Improved Algorithms for Network Topology Discovery -- Using Simple Per-Hop Capacity Metrics to Discover Link Layer Network Topology -- Revisiting Internet AS-Level Topology Discovery -- Section 5: Wireless Network Measurements -- Application, Network and Link Layer Measurements of Streaming Video over a Wireless Campus Network -- Measurement Based Analysis of the Handover in a WLAN MIPv6 Scenario -- Section 6: Monitoring Facilities -- A Distributed Passive Measurement Infrastructure -- lambdaMON - A Passive Monitoring Facility for DWDM Optical Networks -- Section 7: Routing and Traffic Engineering Measurements -- Internet Routing Policies and Round-Trip-Times -- Traffic Matrix Reloaded: Impact of Routing Changes -- Some Observations of Internet Stream Lifetimes -- Section 8: Spectroscopy and Bandwidth Estimation -- Spectroscopy of Traceroute Delays -- Measuring Bandwidth Between PlanetLab Nodes -- Comparison of Public End-to-End Bandwidth Estimation Tools on High-Speed Links -- Section 9: Poster Session -- Traffic Classification Using aStatistical Approach -- Self-Learning IP Traffic Classification Based on Statistical Flow Characteristics -- Measured Comparative Performance of TCP Stacks -- Applying Principles of Active Available Bandwidth Algorithms to Passive TCP Traces -- A Network Processor Based Passive Measurement Node -- A Merged Inline Measurement Method for Capacity and Available Bandwidth -- Hopcount and E2E Delay: IPv6 Versus IPv4 -- Scalable Coordination Techniques for Distributed Network Monitoring -- Evaluating the Accuracy of Captured Snapshots by Peer-to-Peer Crawlers -- HOTS: An OWAMP-Compliant Hardware Packet Timestamper -- Practical Passive Lossy Link Inference -- Merging Network Measurement with Data Transport.

Welcometothe6thInternationalWorkshoponPassiveandActiveMeasurement, held in Boston, Massuchusetts. PAM 2005 was organized by Boston University, with ?nancial support from Endace Measurement Systems and Intel. PAM continues to grow and mature as a venue for research in all aspects of Internet measurement. This trend is being driven by increasing interest and activity in the ?eld of Internet measurement. To accommodate the increasing interest in PAM, this year the workshop added a Steering Committee, whose members will rotate, to provide continuity and oversight of the PAM workshop series. PAMplaysaspecialroleinthemeasurementcommunity. Itemphasizespr- matic, relevant research in the area of network and Internet measurement. Its focus re?ects the increasing understanding that measurement is critical to e?- tive engineering of the Internet's components. This is clearly a valuable role, as evidenced by the yearly increases in the number of submissions, interest in, and attendance at PAM. PAM received 84 submissions this year. Each paper was reviewed by three or four Program Committee (PC) members during the ?rst round. Papers that received con?icting scores were further reviewed by additional PC members or external reviewers (typically two). After all reviews were received, each paper with con?icting scores was discussed extensively by its reviewers, until a c- sensus was reached. The PC placed particular emphasis on selecting papers that were fresh and exciting research contributions. Also, strong preference was given to papers that included validation results based on real measurements.