Nice blog post on approximate quantiles by the guys behind Druid at Metamarkets. The basic technique they use is the histogram proposed by Ben-Haim & Tom-Tov for their Streaming Parallel Decision Tree.
The Art of Approximating Distributions: Histograms and Quantiles at Scale
22 September 2013 by Gianmarco De Francisci Morales
- My Tweets
Archives
Categories
Tags
academia action algorithm Barcelona big data business change character cheers cloud comic computer create data base data mining data science definition destiny dilbert distributed systems economy engineer exchange Facebook failure fear freedom Google graph growth GSoC habit hadoop happiness intelligence Italy japanese life machine learning Mac OS X mapreduce memory news performance phd phdcomics philosophy pig privacy problem procrastination programming proposal quote rant research S4 scalability science sloth SNOW software Storm stream processing syntactic sugar thesis thought time Twitter USA word work world xkcd YahooGravatar
Meta
Kenkyuu 
Leave a Reply