Summary:

The requirement for a big data platform that scales to support massive amounts of data and traffic is fast becoming a must-have tool inside every rapidly growing business. But what is the right reference architecture for a big data stack?

puzzle pic 2

The right architectural stack might look something like this: Ubuntu, Hadoop Distributed File System (HDFS), MapReduce, Cassandra or HBase, Hive, Flume, JDBC and ODBC drivers, Hue, Pig, Oozie, Avro, and Zookeeper, as well as some Chef configuration management tools. The list goes on and on.

How does an IT shop modify and integrate these different software components and hardware into a single big data solution? Overall, IT pros need to understand what problem they are trying to solve, be ready to pick the best components that fit together, and look for vendors that are pulling pieces of the stack together as a solution, taking much of the pain and time out of the integration work.

For big data to succeed in mainstream enterprises, it must be as easy to install and use as a Microsoft Excel spreadsheet.

Join GigaOM Pro and our sponsor Canonical for “Designing for big data: the new architectural stack,” a free analyst roundtable webinar on Wednesday, March 14, 2012, at 10 a.m. PST.

Our panel of experts includes:

Some of the topics we will discuss include:

  • What are the challenges with today’s big data tools and technologies?
  • How are companies using these today?
  • What are the options moving forward?

Claim your spot for this free analyst roundtable webinar. Register today.

Photo courtesy Flickr user andrewmalone

You’re subscribed! If you like, you can update your settings

Comments have been disabled for this post