Campus Problems SolutionsGroup 7 Breakout Results.ppt

11 April 2000 Campus Problems SolutionsGroup 7 Breakout Results Gang of 10 Recorded by Matt Zekauskas Gang of 10 Matt Zekauskas Guy Almes Steve Corbato Joe Perches, Spirent Erik Plesset, Spirent John Streck, NC State Kevin Walsh, SDSC Ronn Ritke, NLANR Joe St Sauver, Oregon Claudia de Luna, JPL Campus Understanding Measurement box at edge Spend month or two with mobile box, checking throughput/loss/.. from every point. Develop a baseline to get a complete picture of the campus: map the campus networks To divide and conquer Measurement infrastructure at all boundaries Automatically fire off tests, given two end points Result “problem is there!” Other tools/metrics Routing db / route views Mtrace on boxes “mperf”? Passive monitoring w/in Abilene AS/AS, top talkers NTP everywhere! If GPS, get good NTP distribution Allow correlation among campuses To work on end-systems Find top 10 systems on campus, and survey ‘em. Running large windows? Recent/decent TCP w/options enabled (SACK) End-systems status Clickable link that would tell you state of host? Applet to run on system for tests End Systems If updating end system (to change TCP parameters) add other things at the same time: IPv6 stacks, SSM (IGRPv3) To work on Microsoft Systems Clickable .REG – change all the defaults! (How lucky do you feel?) Group 7 Breakout: Campus Problems and Roadblocks