Timo-Pekka Heikkinen TKK Networking laboratory Supervisor: Prof. Raimo Kantola Instructor: Lic. Tech. Marko Luoma “Measuring the performance of an active network measurement platform”
Master’s Thesis Presentation – Timo Heikkinen « » Contents ›Theory part »Terms and concepts ›Practical part »Brix system ›Results ›Conclusions ›Future work
Master’s Thesis Presentation – Timo Heikkinen « » Theory part ›Active measurements in data communication networks ›Active vs. passive measurements ›Types of active measurement mechanisms and methods ›Difficulties in making active measurements »Accuracy of measurements
Master’s Thesis Presentation – Timo Heikkinen « » Terms and concepts 1 ›Probe »An artificial packet containing information relevant to the measurement (e.g. timestamp, seq. # etc.) ›One-way latency (delay) »Time it takes for a probe to travel from point A to point B in a network ›Round-trip delay »Time it takes for a probe to travel from A to B and back to A ›Delay jitter »The amount of variation in the measured delay ›Packet loss »A packet is declared lost when it has failed to arrive to its destination in a certain amount of time
Master’s Thesis Presentation – Timo Heikkinen « » ›Types of delay »Processing delay ·Time it takes for a router to process a packet »Transmission delay ·Time it takes for a router to push a packet to the link »Propagation delay ·Time it takes for a signal to travel across the physical medium »Queuing delay ·Time the packet spends inside routers’ queues Terms and concepts 2
Master’s Thesis Presentation – Timo Heikkinen « » Practical part ›Measuring the performance of an active measurement system (Brix Networks) ›Two test cases / network environments »Simple network topology ·Comparison of the Brix system versus an well-known accurate measurement system (AX4000) »(Relatively) Complex network topology ·Is Brix able to detect certain events in the network? ›Overall, how does the Brix system perform?
Master’s Thesis Presentation – Timo Heikkinen « » Devices Brix 100 Verifier Brix 1000 Verifier » Synchronization › Brix 1000 GPS-receiver · Acted as an NTP-server › Brix 100’s NTP synched › NTP synch in the complex case · No need for GPS because RTT was measured
Master’s Thesis Presentation – Timo Heikkinen « » Brix Architecture Operations Center GUI, Subscriber portal GUI Verifier Consolidator Collector Local Registry Network Registry Test BrixWorx
Master’s Thesis Presentation – Timo Heikkinen « » Brix Reporting Data from the Brix system
Master’s Thesis Presentation – Timo Heikkinen « » Results (1/4), case 1 Cumulative probability distributions of all tested devices when measuring one- way delay. Delay shown with and without NTP-offset correction
Master’s Thesis Presentation – Timo Heikkinen « » Results (2/4), case 1 Measured one-way delay between Brix 1000 and Brix 100 verifiers
Master’s Thesis Presentation – Timo Heikkinen « » Results (3/4), case 2 Effects of high load on the delay measured by the Brix system
Master’s Thesis Presentation – Timo Heikkinen « » Results (4/4), case 2 Core router failure as reported by the Brix system.
Master’s Thesis Presentation – Timo Heikkinen « » Conclusions (1/2), case 1 ›More time should have been spent on designing the test setup ›Brix 100’s are not accurate enough to measure sub-millisecond one-way delays »Clock instability big issue »Accuracy ~940 μs, ~120 μs with NTP corr. ›Brix 1000’s are accurate when using GPS synchronization »Clocks stabile enough »Accuracy ~50 μs, ~10 μs with NTP corr. ›AX4000 accuracy ~1 microsecond (when using only one clock)
Master’s Thesis Presentation – Timo Heikkinen « » Conclusions (2/2), case 2 ›Test parameters should have been selected more carefully ›Measurement reports are lost too often ›Congestion and node failures can be detected with Brix ›Short link failures were undetected
Master’s Thesis Presentation – Timo Heikkinen « » Future work ›Case 1 should be done in a more complex network environment ›Case 2 should be done with the same equipment as used in case 1 ›GPS synchronized Brix 1000’s ›NTP offset measurement