September 9, Wireless Internet Performance Research Carey Williamson iCORE Professor Department of Computer Science University of Calgary
September 9, Internet Protocol Stack r Application: supporting network applications and end-user services m FTP, SMTP, HTTP, DNS, NTP r Transport: end to end data transfer m TCP, UDP r Network: routing of datagrams from source to destination m IPv4, IPv6, BGP, RIP, routing protocols r Data Link: hop by hop frames, channel access, flow/error control m PPP, Ethernet, IEEE b r Physical: raw transmission of bits Application Transport Network Data Link Physical
September 9, The Wireless Web r The emergence and convergence of these technologies enable the “wireless Web” m the wireless classroom m the wireless workplace m the wireless home r My iCORE mandate: design, build, test, and evaluate wireless Web infrastructures r Holy grail: “anything, anytime, anywhere” access to information (when we want it, of course!)
September 9, Research Interests r Wireless Internet Technologies r Web Performance r Network Traffic Measurement r Workload Characterization r Traffic Modeling r Network Simulation r Network Emulation
September 9, Wireless Internet Technologies r Mobile devices (e.g., notebooks, laptops, PDAs, cell phones, wearable computers) r Wireless network access m Bluetooth (1 Mbps, up to 3 meters) m IEEE b (11 Mbps, up to 100 meters) m IEEE a (55 Mbps, up to 20 meters) r Operating modes: m Infrastructure mode (access point) m Ad hoc mode r Classroom area networks (CRAN)
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Example: r Multi-hop “ad hoc” networking Carey Gwen
September 9, Web Performance r Explore techniques to improve the performance and scalability of the Web r Examples: m Clustered Web servers m Load balancing policies m Web prefetching strategies m Web proxy caching architectures m Improvements to HTTP and TCP protocols
September 9, Network Traffic Measurement r Collect and analyze packet-level traces from a live network
September 9, Network Traffic Measurement r Collect and analyze packet-level traces from a live network, using special equipment
September 9, Network Traffic Measurement r Collect and analyze packet-level traces from a live network, using special equipment r Process traces, statistical analysis r Diagnose performance problems (network, protocol, application)
September 9, Example Trace > TCP : win: 5840 S > TCP : win: 5792 SA > TCP : win: 5840 A > TCP : win: 5840 PA > TCP : win: 5792 A > TCP : win: 5792 A > TCP : win: 5792 A > TCP : win: 8688 A > TCP : win: A > TCP : win: 5792 PA > TCP : win: 5792 A > TCP : win: A > TCP : win: 5792 A > TCP : win: 5792 A > TCP : win: 5792 A > TCP : win: A > TCP : win: A > TCP : win: 5792 FPA > TCP : win: A > TCP : win: FA > TCP : win: 5792 A
September 9, Time SeqNum X + Key: X Data Packet + Ack Packet X X X X X X X X X X X X X
September 9, Workload Characterization r Try to understand the salient features of network, protocol, application, and user behaviour on the Internet r Example: Web server workloads [Arlitt96] m Zipf-like document referencing behaviour m Lots of “one-time” referencing of documents m Heavy-tailed file size distributions m Self-similar network traffic profile
September 9, Traffic Modeling r Construct programs and statistical models that capture the empirically-observed network traffic behaviours r Allows flexible, controlled, repeatable generation of workloads for experiments r Examples: m Web client workload model m MPEG compressed video model m Self-similar Ethernet LAN traffic model m WebTraff GUI: Web proxy workload generator
September 9, Network Simulation r Use computer simulation to study the packet-level behaviour of the Internet, its protocols, its applications, and its users r Examples: m Improving Web performance over ADSL m Understanding the effects of user mobility on Mobile IP routing and protocol performance m Studying the design, scalability, and performance of Web server and Web proxy caching architectures
September 9, Network Emulation r A hybrid performance evaluation methodology that combines simulation and experimental implementation r A simulator that “talks back” (IP packets) r Examples: m Web server benchmarking m Wide Area Network (WAN) emulation m Web proxy cache performance m Distributed applications (Internet games)
September 9, Summary r Wireless Internet Performance Lab (UofC) r Experimental Laboratory for Internet Systems and Applications (UofS/UofC,CFI) r iCORE Research Team: m Five full-time research staff (Web, perf. eval., simulation, wireless, traffic modeling, network measurement) plus 7 graduate students r Research Collaborations: m UofC, UofA, UofS, TRLabs, CS/ECE m HP, Telus Mobility, SaskTel, Sun, Nortel… r Do cool, “hands on”, industrially-relevant, applied, practical, and exciting stuff!!