Towards continuous aggregate analytics

A session at Hadoop User Group UK

Acunu Analytics helps solve the problem of computing simple analytics
over large datasets, incrementally, and in real-time. For many workloads, it answers queries in milliseconds rather than the minutes needed by Hadoop and mapreduce. It is also write-friendly, supports thousands of writes per second, many times more than a relational store can handle. Like Haodop, it scales up to hundreds of millions of events and metrics stored durably to disk across multiple nodes and datacenters. It provides a simple yet powerful set of aggregation functions, including count, sum, average, variance, count distinct, and top-k. Using these primitives, you can do things like:

  • Tracking trending topics (e.g. top-10 by category)
  • Network monitoring (e.g. looking for infected nodes)
  • Analytics dashboards (e.g. tracking ad impressions)
  • Operational intelligence (e.g. infrastructure monitoring)
  • and more...

Storage is handled by Cassandra, so you get all of the benefits of Cassandra’s reliability, scalability, and load balancing. You can submit events and run queries through a simple HTTP REST interface. A Flume plugin is also provided so you can run analytics directly over log files.

About the speaker

This person is speaking at this event.
Tom Wilkie

Founder, CTO & VP Engineering @acunu I tweet about my company (Acunu) and my RepRap. bio from Twitter

Sign in to add slides, notes or videos to this session

Tell your friends!

When

Date Tue 22nd May 2012

Short URL

lanyrd.com/srtrm

Official session page

www.acunu.com

View the schedule

Share

See something wrong?

Report an issue with this session