internet measurement (presentation)

1. Internet Traffic Measurement Amir Hossein Mandegar Dr. B. Akbari September 2011 Computer Engineering and IT Department

2. Internet Traffic Measurement 2 Outline Why ? Tools Challenges Applications

3. Internet Traffic Measurement 3 Why Measure ? Although Internet works, it is far from being ideal Measurements of various aspects of it will: Help us to better understand why it works the way it does Help us to diagnose known problems and lead us one step closer to their solutions Help us to design new features that the Internet should provide to enable next-generation application requirements Simply put, Internet Measurements is key to the design of the next-generation Internet

4. Internet Traffic Measurement 4 What to Measure ? Physical Properties Devices (routers, NAT boxes, firewalls, switches), Links (wired, wireless) Topology Properties Various levels Autonomous Systems (AS), Points of Presence (PoP), Routers, Interfaces Traffic Properties Delays (Transmission, Propagation, Queuing, Processing etc.), Losses, Throughput, Jitter

6. Internet Traffic Measurement 6 Measurement Tools Active Measurement Passive Measurement Fused/Combined Measurement Inference Measurement Others

7. Internet Traffic Measurement 7 Active Measurement Tools Methods that involve adding traffic to the network for the purposes of measurement Ping: Sends ICMP ECHO_REQUEST and captures ECHO_REPLY Useful for measuring RTTs Only sender needs to be under experiment control Traceroute: Ping nodes with incremental TTL from one Find Hops between source and destination node Only sender needs to be under experiment control There is some care about the result

8. Internet Traffic Measurement 8 Active Measurement Load Active Measurement inserts considerable load on network links if attempting a large-scale topology discovery Optimizations reduce this load considerably If single source is used, instead of going from source to destination, a better approach is to retrace from destination to source If multiple sources and multiple destinations are used, sharing information among these would bring download considerably

9. Internet Traffic Measurement 9 Passive Measurement Capture traffic generated by other users and applications Routeview collects BGP views (routing tables) from a large set of ASes OSPF LSAs can processed to generate router graphs within an AS

10. Internet Traffic Measurement 10 Fused Measurement Active Measurement + Passive Measurement + Fused Measurement mixed method to use benefits of each method. Less Traffic Load Less Required Storage More information for decision

11. Internet Traffic Measurement 11 Inference Measurment Measurement Limitations Direct access impossible Topology and link out of reach How to solve ? End-to-End Measurement Tomography: process of inferring network topology, delays, packet losses etc. using only end-to-end measurements

13. Internet Traffic Measurement 13 Measurement Level Bit & Bytes (useless) Packets (High volume) High volume but valuable data for offline & generate traffic Flow (limited to flows) Protocols (Specific Protocol) TCP, UDP, IP, RTP, .. Application (Known applications) HTTP, DNS, MAIL, VOIP, ... SNMP

14. Internet Traffic Measurement 14 Hidden Pieces - Middleboxes Firewalls provide security Traffic Shapers assist in traffic management Proxies improve performance NAT boxes utilize IP address space efficiently Each of these impedes visibility of network components. Example: firewalls may block active probing requests NATs hide away the no. of hosts and the structure of the network on the other side

15. Internet Traffic Measurement 15 Trace Challenges Payload Privacy Encryption Sanitization Reduction Techniques (OC-48, 50% utilization, 155MBps) TCP: Syn/Fin/Rst Uni-direction Analysis Sampling Techniques

17. Internet Traffic Measurement 17 Bandwidth Measurment Bandwidth Measurement Amount of data the network can transmit per unit time Three kinds of bandwidth capacity: max throughput a link can sustain, available bandwidth: capacity used bandwidth bulk transfer capacity: rate that a new single long-lived TCP connection would obtain over a path

18. Internet Traffic Measurement 18 Path Provisioning by Traceroute Suppose the path between A and D is to be determined using traceroute A X Y D B C

19. Internet Traffic Measurement 19 Traceroute Process A X Y D B C Dest = D TTL = 1 B: time exceeded

20. Internet Traffic Measurement 20 Traceroute Process A X Y D B C Dest = D TTL = 2 C: time exceeded

21. Internet Traffic Measurement 21 Traceroute Process A X Y D B C Dest = D TTL = 3 D: echo reply

22. Internet Traffic Measurement 22 Traceroute issues Path Asymmetry (Destination -> Source need not retrace Source -> Destination) Unstable Paths and False Edges Aliases Measurement Load

23. Internet Traffic Measurement 23 Unstable Paths and False Edges Inferred path: A -> B -> Y A X Y D B C Dest = D TTL = 1 B: time exceeded Dest = D TTL = 2 Y: time exceeded

24. Internet Traffic Measurement 24 Aliases IP addresses are for interfaces and not routers Routers typically have many interfaces, each with its own IP address IP addresses of all the router interfaces are aliases Traceroute results require resolution of aliases if they are to be used for topology building

25. Internet Traffic Measurement 25 Video Quality Measurment Subjective Evaluation Variety cause of user experience Need many user to vote Objective Evaluation Data Metrics (PSNR) Picture Metrics (MPQM ,SSIM) Packet/Bit stream Metrics (Loss) Full/Reduced/Non Reference Visible/Invisible distortion

26. Internet Traffic Measurement 26 Video Quality Measurment GOP Frames (P, I, B) Slices

27. Internet Traffic Measurement 27 Video Quality Measurment Stream should be packetized TS(DVB) vs RTP(IPTV) Passive Measurement Deep Packet Inspection Resilience and Error correction Quality Estimation (Fuzzy, NN, ...)

28. Thank You!