Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
The Campus as key to Internet2 Engineering San Diego Guy Almes <[email protected]> 4 December 2000 Outline of Talk A Internet2 Engineering Objectives A The Logic of End-to-End Performance • Our Aspirations • Threats to these Aspirations • Promising Approaches to Success A The Internet2 End-to-End Performance Initiative Internet2 Engineering Objectives A Provide our universities with superlative networking: • Performance • Functionality • Understanding A Make superlative networking strategic for university research and education The End to End Challenge A Support advanced networking end to end A Performance • 100 Mb/s across the country normative • several multiples possible in some cases A Functionality • Multicast • Quality of Service • IPv6 • Measurements What are our Aspirations? A A Candidate Answer #1: Switched 100BaseT + Well-provisioned Internet2 networking ® 80 Mb/s But user expectations and experiences vary widely What are our Aspirations? A A Candidate Answer #2: Lower user expectations and minimize complaining phone calls There is a certain appeal I suppose... What are our Aspirations? A A Candidate Answer #3: Raise expectations, encourage aggressive use, deliver on performance/functionality to key constituencies. Not the easy way, but necessary for success Why should we Care? A Advanced faculty needs: • Effective access to remote facility: quickly move large datasets. PPDG: 400 Mb/s to CERN by 2003. • Interactive access: video or control or VoIP. Very low loss/jitter. A We (in several senses) need to deliver the goods. Why should we Care? A "We" as the university community. A "We" as campus networking specialists. A "We" as networking professionals. A "We" as the (broad) Internet2 project. A Low aspirations are dangerous to us. Abilene core November 2000 Seattle New York Cleveland Indianapolis Sacramento Washington Denver Kansas City Los Angeles Atlanta Houston Abilene Connections by (roughly) YE 2000 International Peering Seattle CA*net3, (AARnet) STAR TAP APAN/TransPAC, CA*net3, IUCC, NORDUnet, RENATER, REUNA, SURFnet, SingAREN, SINET, TAnet2 CERnet, (HARnet) OC12 Sunnyvale New York DANTE*, JANET, NORDUnet, SURFnet CA*net3 (SINET) Los Angeles SingAREN, SINET San Diego CUDI El Paso (CUDI) OC3-12 Miami (REUNA, RNP2, RETINA) The Current Situation A We have a combined Internet2 infrastructure of considerable capacity • examples of 240 Mb/s flows A End to end performance varies widely • but 40 Mb/s flows not always predictable • users don't know what their expectations should be A Note the mismatch The Current Situation A We have a combined Internet2 infrastructure of considerable capacity • examples of 240 Mb/s flows A End to end performance varies widely • but 40 Mb/s flows not always predictable • users don't know what their expectations should be A Note the mismatch Threats to End to End Performance A A BW = C x packet-size / ( delay x sqrt(packet-loss )) (Mathis, Semke, Mahdavi, and Ott, CCR, July 1997) Context: • Network capacity • Geographical distance • Aggressive application Threats to End to End Performance A Network Path • • A local / department / campus gigaPoP / backbone / exchange points Host problems • OS / TCP • Hardware: NIC, CPU, memory, bus • Application Threats to End to End Performance A Fiber problems • dirty fiber • dim lighting • 'not quite right' connectors Threats to End to End Performance A Fiber problems A Switches • horsepower • full vs half-duplex • auto-sense 10/100 • head-of-line blocking Threats to End to End Performance A Fiber problems A Switches A Inadvertently stingy provisioning • mostly communication • happens also in international settings Threats to End to End Performance A Fiber problems A Switches A Inadvertently stingy provisioning A Wrong Routing • asymmetric • best use of Internet2 • distance Threats to End to End Performance A Fiber problems A Switches A Inadvertently stingy provisioning A Wrong Routing A Host issues • NIC • OS / TCP stack • CPU Perverse Result A A 'Users' think the network is congested or that the Internet2 infrastructure cannot help them 'Planners' think the network is underutilized, no further investment needed, and users don't need high performance networks Promising Approaches A Work with key motivated users A 'Shining a flashlight' on the problem A Measurements A Divide-and-Conquer A Understanding Application Behavior A Getting it right the first time Active Measurements within Abilene Surveyors with: Active delay/loss measurements Ad hoc throughput tests Application to Performance Debugging Application to Performance Debugging Divide and Conquer A A Systematically identify/isolate the network segment at fault Can we make this systematic and (eventually) automated? End to End Advanced Functionality A Multicast A IPv6 A QoS Internet2 End-to-End Performance Initiative A A A Distributed measurement infrastructure Teams of performance analysis specialists (PERTs) Dissemination of best practices Defining End-to-End Success Metrics A Identify core applications / services • high-performance TCP • VoIP / videoconferencing • pervasive native IP multicast A Scope • How pervasive is it supported across the campus? A Timeliness • When are these metrics achieved? Anticipated Partners A NLANR A Web100 A Abilene partners A Leading campuses and gigaPoPs A Internet2 corporate partners Initiative Phases A 1st Gear • Preparation, planning, early experiments A 2nd Gear: Early Adopters Phase • Partner with 10-15 selected campus • Develop PERTs, Measurement Infrastructure, etc. • Build tools, resources, and best practices • Expect RFP in late January 2001 A 3rd Gear: Dissemination • Increasingly pervasive PERTs, infrastructure Creating Internet2 Value A A A A Build the infrastructure together Make end-to-end performance and advanced functionality routine Identify and connect valuable resources for our faculty and students Have fun