This lecture will be given today at Tel-Aviv University, school of
Comptuer Science, Schreiber building.
Time: 17:00-20:00 
Room: 006
Map: http://www2.tau.ac.il/map/unimapl1.asp



Hadoop at Yahoo! State of technology today and future development. 

Talk Abstract
As the most visited site on the Internet Yahoo! has to build products
that scale to thousands of servers. Adoption of Behavioral Targeting &
WebAnalytics into the production cycle forced web companies to build
expensive data pipelines with data warehouse and myriad of compute
clusters. Then, grid computing that for a long time strove to find its
place in the industry turn to be the right tool for the job. In 2006
Yahoo! started with open-source project called Hadoop with a goal to
build stable & scalable grid solution that include distributed file
system (HDFS) and Map Reduce framework. Today Y! Grid Technology Team
completes migration of all Y! data driven businesses to Hadoop. The
project quickly gained momentum in open source community attracting
dozens of contributors. Every month we host Hadoop User Group at Yahoo!
that became a meeting place for Hadoop developers and users from
Facebook, IBM, Google, Ebay and many Silicon Valley startups. 

In my presentation I will talk about how Web Giants use data,
technologies that were used so far, and how Hadoop helps to streamline
product development and R&D. We will also cover the current state of
Hadoop technology and next year Roadmap.

BIO.
Michael Pilip is Sr. Product Manager in Y! Cloud Computing & Data
Infrastructure division responsible for Hadoop development and Y!
products migration to grid. Before joining Hadoop Michael worked on Y!
analytical data pipeline that brings data from dozens of thousands web
servers around the globe to Data Marts and Behavioral Targeting
applications. Prior to turning to data business Michael was a lead
developer in Y! Games building the biggst on-line games portal in the
World.

Agenda:
How web giants use user-generated data (WebAnalytics, Behavioral
Targeting, Reporting, R&D) Data Pipelines architecture (Data Collection,
ETL, Data Warehousing, Aggregations, DataMarts, AdServing Systems)
Hadoop to the rescue Hadoop at Yahoo! (development and cluster
operations) Hadoop open-source projects and major trends.
Q&A     

================================================================To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]

Reply via email to