BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:San Fermin: Aggregating Large Data Sets using Dynamic Binomial Tre
 es - Justin Cappos\, University of Arizona
DTSTART:20070621T153000Z
DTEND:20070621T163000Z
UID:TALK7632@talks.cam.ac.uk
CONTACT:Wenjun Hu
DESCRIPTION:Content aggregation is an important sub-problem in distributed
  monitoring\, distributed database queries\, and software debugging. In th
 is problem there are a large number of systems that have information and t
 he requester is not interested in the result from each individual machine\
 , but rather the aggregated results from all machines. Current solutions t
 o this problem have looked at the case where the aggregate data is\nsmall 
 (typically only a few bytes) and typically aggregate data by running a mul
 ticast tree in reverse.\n\nThis talk describes a novel algorithm called Sa
 n Fermin used to aggregate large data sets. San Fermin returns the answer 
 from more nodes\, computes the result faster\, and has better scalability 
 than\nexisting solutions. Our evaluation explores different aggregation te
 chniques using mathematical modeling\, simulation\, and deployment of a pr
 ototype on PlanetLab. Evaluation shows that San Fermin is scalable as eith
 er the number of nodes or the data size increases. San Fermin is also amaz
 ingly resilient to failures\, so that when 10% of the nodes fail during ag
 gregation it still returns the answer from over 97% of the nodes.\n\nBio: 
 \n\nJustin Cappos is currently working on his Ph. D. at the University of 
 Arizona with John Hartman and Beichuan Zhang. His research is focused on i
 mproving the security and efficiency of real world networks of\ncomputer s
 ystems. He has lead a number of projects including Stork.\n
LOCATION:Lecture Theatre 2\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR