Get short, timely messages from CloudComputing@CMU.

Twitter is a rich source of instantly updated information. It's easy to stay updated on an incredibly wide variety of topics. Join today and follow @CloudAtCMU.

Get updates via SMS by texting follow CloudAtCMU to 40404 in the United States
Codes for other countries

Two-way (sending and receiving) short codes:
Country Code For customers of
Australia
  • 0198089488 Telstra
Canada
  • 21212 (any)
United Kingdom
  • 86444 Vodafone, Orange, 3, O2
Indonesia
  • 89887 AXIS, 3, Telkomsel
Ireland
  • 51210 O2
India
  • 53000 Bharti Airtel, Videocon
Jordan
  • 90903 Zain
New Zealand
  • 8987 Vodafone, Telecom NZ
United States
  • 40404 (any)

CloudAtCMU

  1. IBM Picks Hadoop To Analyze Large Data Volumes http://bit.ly/95ToY9 (via @InformationWeek)
  2. "Experiences with a CANoe-based Fault Injection Framework for AUTOSAR" accepted to DSN 2010. Failure diagnosis for automotive systems.
  3. Work on Ganesha, black-box diagnosis for Hadoop, makes it into ACM SIGMETRICS Performance Evaluation Review. http://bit.ly/bamOjr
  4. Similar work in Master's thesis by @eugeneiim on Map-Reduce mobile cluster http://bit.ly/bw3AvW. RT @awadallah: http://bit.ly/9KFdq6 (@om)
  5. Great collection, variety of uses: Mapreduce and Hadoop algorithms in academic papers http://is.gd/8g9ry (via @yarapavan and @alisohani)
  6. Our work on "Visual Log-based Causal Tracing for Performance Debugging of MapReduce" to appear at ICDCS 2010 #hadoop #cloudcomputing
  7. Interesting numbers on data centers for the movies. RT @datacenter Avatar, Hollywood & the Data Center http://bit.ly/dmHLMf
  8. Our analysis of production traces from Yahoo's M45 cluster on Dr. Dobbs: http://www.ddj.com/222400160?cid=RSSfeed_DDJ_All #hadoop
  9. Traces from a Google production cluster, now available for researchers to analyze. http://bit.ly/4qwIvM #cloudcomputing
  10. Our work on "Analysis of Traces from a Production MapReduce Cluster" to appear at IEEE/ACM CCGrid 2010 conference #hadoop #cloudcomputing
  11. Blog article on Jeff Dean's talk on Google's infrastructure: What goes wrong in a typical cluster's first year http://bit.ly/4ySjQo
  12. Our work on "Kahuna: Problem Diagnosis for MapReduce-Based Cloud Computing Environments" to appear at IEEE/IFIP NOMS 2010 conference
  13. Our work on ""Hardware Performance Counter-Based Problem Diagnosis for e-Commerce Systems" to appear at IEEE/IFIP NOMS 2010 conference
  14. Our work on black-box failure diagnosis for parallel file-systems (PVFS, Lustre) was accepted for USENIX FAST 2010! See you there.
  15. Awesome -- @tudor_dumitras wins the 2009 John Vlissides award for PhD students at OOPSLA 2009! Picture: http://bit.ly/4iOhh
  16. Tech report CMU-CS-09-143: @tanjiaqi's MS thesis, log-based approach to characterizing, diagnosing & visualizing MapReduce behavior #hadoop
  17. At Portugal Telecom in Lisbon last week. Interest in and lots of Q&A about our log-analysis and diagnosis. #cloudcomputing #debugging
  18. Imago, our zero-downtime, large-scale upgrades approach, presented by @tudor_dumitras at the SEI, CMU #upgrades #cloudcomputing
  19. Our black-box, syscall-based diagnosis for PVFS, at HotDep 2009 (at DSN, in Portugal) presented by Mike Kasick #cloudcomputing #hotdep
  20. Why "Ganesha"? Ganesha is the Hindu God who is the remover of obstacles (problems in MapReduce). Ganesha also has the head of an elephant