Versions Compared

Old Version 94

changes.mady.by.user Les Cottrell

Saved on

compared with

New Version 95

changes.mady.by.user Les Cottrell

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  1. Deployment: Faran
    1. Provide a program API: Faran, Shahryar (dropped)
      Refine ##Refine the list in Sites.txt by writing a script that will run TULIP HTTP requests to ping to see which landmarks are not responding (agreed 11/16/06, do we really need it, if so can we reduce its priorityit it is (for me) just as easy to edit the file via an editor at SLAC): Faran
    2. Provide opotion so that Sites.txt can be modified from TULIP. (agreed 11/16/06, currently the master is kept at SLAC and one has to logon to SLAC to modify it): Faran
    3. Ignore comments in mid line (agreed 11/30/06, is it done 1/31/07): Faran
    4. Look into Mathematics of Singular Vector Decomposition technique. (agreed 11/16/06): Faran
    5. SLAC account for Faran (Faran need to report if password changed): Faran
  2. Visualization
    1. Make tables sortable (working on 5/25/06, re-agreed 8/3/06, re-requested 8/21/06, done for the RTTs 8/22/06, can we do with buttons for all columns): Faran
    2. Can we add AS's to routes: Yee, Faran
    3. Integrate with Google maps: Faran
    4. When zooming maintain size of dots:  Faran
  3. Landmarks
    1. Get more lanmarks for TULIP:
      1. Sent email Latin America (5/7/06, 11/25/06): Les
        1. Follow up on positive response from CUDI Mexico (reminded 1/12/07, it is there but still await the URL, sent reminder 1/27/07): Les
      2. China (sent email to ctsnet 7/26/06, reminded 12/6/06, 1/27/07): Les
      3. Try and get sites in Africa (Les sent email 11/28/06): Les
      4. COMTEC, Japan (Faran send email reminder 5/25/06, agreed to send another reminder 6/7/06, Umar will check with Farooq 11/16/06): Faran, Farooq
      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
    2. Add NetDiGi route servers (asked for 8/30/06): Faran
    3. Provide tool to geolocate all PingER sites, analyze results to see what landmarks never respond; Faran
    4. Problem Landmarks:
      1. pinger.ntc.net.pk denies accesss (email to Azher & Faran 7/25/06, Azher responded that he is checking 7/27/06, email sent again 8/21/06, Azher knows about will send person to repair 8/22/06) see also above for the PingER node at this site: Faran/Azher
      2. Enable landmarks that only work for IP addresses: Faran
  4. 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): Faran
  5. Analysis
    1. Literature search on triangulation, multi-lateration: Faran
    2. Measure landmarks from landmarks: Faran
    3. Document Faran's algorithm, try new algorithm with optimized alpha that does not require destination to be inside triangle: Faran, Umar
    4. Analyse the effect of passing IP addresses to landmarks rather than names: Faran
    5. Compare Faran's algorithm with Bamba's, Octant & simple triangulation, GeoIP tools (in progress 2/1/07)
      1. Contact Bamba to find out if http://planetlab-01.ipv6.lip6.fr:10000/cbg.php works for any sites (email send 4/21/06): Aziz
      2. Confidence estimates
      3. Applied to PingER hosts
      4. Outside triangles, area coverage
      5. Effect/need to vary alpha: Faran
  6. Analysis of Accuracy and applicability
    1. Accuracy vs known locations
    2. Effect of removing, adding landmarks, try for Bolivia:Faran
    3. Comparison with other sets of landmarks & methods (can't do Bamba since it is down), Octant, NetGeo for accuracy coverage etc.
    4. Provide bubble maps of accuracy: Faran
    5. Review difference between Guthrie and TULIP
      1. For US using Octant: Faran
      2. Look at big discrepancies: Faran
      3. Compare with GeoIP: Faran
    6. What area/population is covered by TULIP
    7. Document anomalies  (e.g. root-servers): Faran
    8. Show accuracy by region and country: Faran
    9. Come up with a way to estimate the accuracy/confidence of estimates: Faran
      1. Document in the TULIP web page: Faran

...