Welcome%20to%20the%20Winter%202004%20ROC%20Retreat

1
Welcome to the Winter 2004 ROC Retreat

• Armando Fox and David Patterson

2
About ROC Retreats

• Purpose of semi-annual retreats
• Progress reports/talks from academia and industry
• Exposure/feedback on new ideas or work in
  progress
• Brainstorming in immersive atmosphere
• Industry/visitor feedback, opportunities for
  collaboration
• Skiing
• Logistics
• Web server with retreat talks/papers - thanks to
  Mike Howard and Bob Miller
• Skiing

3
ROC Events

• Aaron Brown, UC Berkeleygt Dr. Aaron Brown, IBM
  Research
• Pete Broadwell, UC Berkeleygt Pete Broadwell,
  M.S., ???
• Soon Mike Chen, UC Berkeleygt Dr. Mike Chen,
  ???
• ROC work recognized in the 2003 Scientific
  American 50

4
Recent Publications (since June 2003)

• Published or to appear
• Ben Ling, Emre Kiciman, Armando Fox Session
  State Beyond Soft State, in NSDI 2004
• Mike Chen, Anthony Accardi, Emre Kiciman, Jim
  Lloyd, Eric Brewer, Armando Fox Path-Based
  Failure and Evolution Management, in NSDI 2004
• George Candea, Steve Zhang, Emre Kiciman, Armando
  Fox, Application-Generic Recovery for Internet
  Middleware, Cluster Computing Journal (special
  issue on Autonomic Computing), summer 2004
• George Candea, James Cutler, Armando Fox,
  Improving Availability with Recursive
  Microreboots A Soft-State System Case Study,
  Performance Evaluation Journal, 56(1-3), March
  2004
• In submission
• George Candea and Armando Fox, Microreboots An
  Application-Generic Recovery Technique for
  Internet Services, submitted to USENIX 2004
• Andy Huang and Armando Fox, Free Recovery A Step
  Towards Self-Managing State, submitted to USENIX
  2004
• Emre Kiciman and Armando Fox, Detecting and
  Localizing Anomalous Behavior to Discover
  Failures in Component-Based Internet
  Services, submitted to USENIX 2004
• Yee-Jiun Song, Jeff Raymakers, Wendy Tobagus,
  Armando Fox. Is MTTR More Important Than MTTF For
  User-Perceived Availability?, submitted to
  DSN-IPDS 2004

5
Preview of some upcoming talks

• Benchmarking
• Evaluating undo human-aware recovery benchmarks
• Benchmarking distributed services
• Including latency data quality in
  performability evaluation of a web-based service
• Making recovery nearly free
• Evaluating the effect of micro-reboots on end
  users
• How cheap recovery simplifies persistent state
  management
• Embracing statistical analysis
• Using statistical learning to detect and localize
  faults in componentized Internet services
• A statistical learning approach to failure
  diagnosis for eBay
• Toward generalized APIs for statistical
  monitoring

6
ROC gt RADS

• Generalize ROC approaches that focus on
  statistical anomaly detection as a way of
  detecting conditions that require response
• Generalize recovery to adaptation
• System is always recovering/always adapting
• Some early examples of this will be featured in
  talks
• Insight statistical pattern recognition provides
  a degree of application-generic failure
  detection nearly-free recovery means we can
  tolerate some false positives
• Kickoff panel this evening

7
Other Highlights

• Poster advertisements before poster session
• Three talks from industrial visitors
• Moises Goldszmidt statistical pattern
  recognition applied to systems management
• Chris Overton modeling large-scale IT systems
• Paul Brett Real-world failures, a systemic view