You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 2 Next »

Triangulate position of an unknown host (see http://www.slac.stanford.edu/comp/net/wan-mon/tulip,Faran Faran expects to be able to devote more effort to this following 25th Jan 2007): Faran, Les

  1. Deployment: Faran
    1. Look into Mathematics of Singular Vector Decomposition technique. (agreed 11/16/06): Faran
  2. Visualization
    1. Provide documentation (in progress 2/27/07, documented maths of selection 3/20/07, needs updating, agreed to update 3/22/07, should it be a wiki? - Umar thinks good idea need Faran view 4/30/07, Faran agrees 5/10/07): Faran
    2. Can we add AS's to routes: Yee, Faran
    3. DiscoveryStatistics =>LandMark Analysis need to identify which Site number is displayed (i.e. from which window): Faran
    4. Sites.txt
      1. Are all landmarks entered?
      2. Need to be able to edit existing sites: Faran
    5. When zooming maintain size of dots (Faran will propose to Umar as a student project 4/10/07, he is concerned about the download time, Asif suggests using NASA images for Java platform, will send link 4/12/07, assigned to Rehan 5/10/07):  Faran, Rehan
  3. Bugs:
  4. Evaluate whether using traceroute data helps (Suggested by Umar & Les 2/26/07, could add 3 more RTTs to minimize from): Faran
  5. Improvements
    1. Allow alpha to change (lot of work, many changes, will improve greatly, changes design, Faran focussing on 4/12/07, need progress)
    2. Look at using PingER historical data to optimize alpha (agreed 4/12/07): Faran
  6. Need a way to display alpha used: Faran
  7. Landmarks
    1. Get more landmarks for TULIP, ideally need at least one landmark in all major developing countries that do not have direct connections to nearby countries.:
      1. China (sent email to ctsnet 7/26/06, reminded 12/6/06, 1/27/07): Les
      2. Try and get sites in Africa (Les sent email 11/28/06): Les

      3. COMTEC, Japan [Done 5/11/07]: Les

      4. HSPC, Hsinchu, Taiwan [Done 5/11/07]: les

      5. Add traceroute/Looking Glass servers in critical areas, if they have a common format (asked for 11/15/06, Les did a manual look at traceroute.org and found about 20 potential landmarks 11/17/06, Faran not clear if one can automate 2/27/07, several are automateable, to prove it Les has added 8, Faran will work on, re-requested 3/26/07 with suggestions, Faran is having problems with blocking after several accesses, will get a list then will try and contact to open up 5/10/07): Faran
      6. Get site in Korea at Ajou university (trying to persuade 5/8/07): Umar
      7. Get site at KHU (server available, problems with permissions to open UDP sockets 5/8/07, Bartelt knows of nothing in the Apache config file that would cause this, may be a priviledges the command line user has more privs than the web server 5/11/07, traceroute server is not responding 5/11/07): Umar
    2. Add NetDiGi route servers (asked for 8/30/06, is able to parse them, will look at adding appropriate hosts 3/22/07): Faran
  8. Apply to PingER node database to check the lat/longs are roughly correct (tried with 300-400 sites but cannnot get results due to fragile Internet at NIIT, will interact with Akbar to make tests at SLAC 8/31/06, exchanged emails on list of PingER hosts to locate 9/3/06, Shahryar sent information on how to access pingtable.pl data): Faran
  9. Analysis
    1. Literature search on triangulation, multi-lateration: Faran

    2. Measure landmarks from landmarks [Done 2/15/07]: Faran

    3. Optimize alpha by landmark (after have new location algorithm working, will try 1:2 for US, 1:3 for Europe, 2:15 Africa, done at NIIT, need to deploy at SLAC 4/10/07) : Faran
    4. Document Faran's algorithm, try new algorithm that does not require destination to be inside triangle: Faran, Umar
    5. Analyse the effect of passing IP addresses to landmarks rather than names: Faran
    6. Look at impact of increasing number of pings for getting accurate min_RTT (Faran measured for a few sites, Les gave ideas on how to extend and analyze for multiple sites, could be a major project 4/4/07, found paper recommending about 10): Faran
    7. Compare with GeoIP Tools

      1. Find a case where TULIP beats GeoIP Tool [Done 4.68.116.16, probably backbone routers too]:

    8. For US compare Faran's algorithm with GeoIP & simple triangulation for PingER hosts(in progress 2/1/07)
      1. Confidence estimates
      2. Applied to PingER hosts
      3. Outside triangles, area coverage
      4. Effect/need to vary alpha: Faran
      5. Look at big discrepancies
  10. Analysis of Accuracy and applicability
    1. Accuracy vs known locations (landmarks & PingER hosts): faran
    2. Provide bubble maps of accuracy: Faran
    3. What area/population is covered by TULIP
    4. Show accuracy by region and country: Faran
    5. Come up with a way to estimate the accuracy/confidence of estimates: Faran
      1. Document in the TULIP web page: Faran
  11. Look at publication (helps advance Faran's career and possibly "best project",  will need more analysis)
    1. Find conference: all
    2. Can we get Cisco interested?: Arshad, Faran
    3. Update documentation: Les, Faran
    4. Read paper at http://research.microsoft.com/%7Epadmanab/papers/sigcomm2001.pdf: Faran
  • No labels