Hadoop at Bloomberg

Posted by:Netsoc Posted on:Mar 23,2015



Bloomberg are coming in to talk about Hadoop, algorithms for processing extremely large data sets! and it’s OPEN SOURCE!

It should be fun and a must for if you’re interested in distributed systems or big data!

They’ll be dinner provided as well (FREE FOOD!)

Remember 6pm in the Maxwell Theatre on Wednesday 25th March!

Bloomberg, the premier data and analytics provider to the financial world, is working on consolidating a number of its core data systems around distributed commodity platforms. HBase is a natural fit in many ways for time series data, but the absolute performance and tight availability tolerances are more restrictive then the general web case.

Our “medium data” problem of dozens of terabytes of data with high reliability and consistently low latency is an interesting one to consider as the capabilities and maturity of the Hadoop ecosystem continue to expand. In this talk, we’ll detail our experiences in driving towards performance goals on high core modern machines, and which techniques have born fruit, from multiple region servers per machine to synchronized garbage collection to secondary region servers, and offer a glimpse of future development.


No description.Please update your profile.


    Friday at 3pm in the Netroom (end of the hall to the left as you come into the Goldsmith)


    March 2015
    M T W T F S S
    « Nov   Sep »