Page History

...

1. Evaluate PCA multivariate method of Crovella: Waqar, Adnan,
   1. Compare PCA results with Holt-Winters/plateau: Adnan
      1. Fetch Data From SLAC, BNL, CALTECH (Pathchirp, Thrulay, Ping): Adnan
      2. Fetch data from CERN
      3. Apply PCA on same data and to get potential alerts: Adnan
      4. Comparison of Run-period results and PCA results: Adnan
      5. Improve event detector: Adnan
      6. Docment findings on PCA: Adnan
2. Forecasting comparison of Active vs. passive data: Fareena, Adnan
   1. Look at eview implementation of ARIMA (QU has copy): Fareena
      1. Extend to more sites/metrics, see need for Integration & higher orders: Farena, Adnan
      2. Try with longer time base: Fareena
      3. Get time estimates : Fareena
   2. Detailed Graphs description: Fareena
   3. Wiki MarkupRegularization \ [Done\]
   4. Look at moodss (http://jfontain.free.fr/statistics.htm): Fareenaunmigrated-wiki-markup
   5. -Send results/documentation to SLAC \ [Done 3/15/07\]: Fareena-
   6. Share documentation with COMSATS: Umar
   7. Review write up, and send changes by FAX to +92-51-9280782: LesAutomate procedures
   8. Define project
   9. Put together presentation for workshop (for Sep 2006): Fareena
   Gather team
3. Automatic event isolation: Adnan, Yee
   1. Review the results we are getting at the moment
   2. Provide email lists to get all reports from all monitors together; Jaredg
4. Extend to closer to real time, especially for loss of connectivity or more frequent measurements