8 Hadoop Big Data Predictions for 2015
December 22, 2014 10:28 PM
Where is Hadoop, the open source Big Data platform, heading in 2015? No doubt, a growing number of businesses will build smart applications atop Hadoop. But the effort will push beyond analytics, according to Forrester Research Inc., which offers up these eight Hadoop predictions for 2015. Image: iStock
1. Believe the Hadoop Hype
Yes, Hadoop is fulfilling its promise for data lakes, traditional and advanced analytics, transactional data, and more. True believers include Wal-Mart, Verizon, Netflix and more. As a result, CIOs who are late to the Hadoop game will finally make the platform a priority in 2015, Forrester predicts. Image: iStock
2. SQL Is Hadoops Killer App
SQL is the primary lingua franca for structured enterprise data and is used by application developers to read and write data to databases, Forrester points out. And, all major BI and analytics tools support SQL Thats why SQL in and on Hadoop has such a strong use in enterprises. Image: iStock
3. Enterprise Software Companies Move In
Hadoop is a general-purpose platform. That means it can store and process any kind or data, Forrester notes. But Hadoop-based data management and governance solutions have gaps. Eager to fill the gaps, well-known names like SAS Institute, Informatica, Teradata, BMC, Protegrity, Talend, IBM, Oracle, Microsoft are launching data analytics, management, and governance tools that run natively in Hadoop and work with enterprise apps. Image: iStock
4. Hadoop Talent Goes Mainstream
Amid Hadoops rapid adoption, businesses and service providers have sometimes suffered from a shortage of Hadoop experts. But the shortage of Hadoop skills will quickly disappear as enterprises turn to their existing application development teams to implement projects such as filling data lakes and developing MapReduce jobs using Java, Forrester predicts. In theory, that means CIOs will increasingly leverage in-house Hadoop talent rather than high-priced consultants. Image: iStock
5. Hadoop Clusters in the Cloud
Hadoop clusters require storage, compute, and network resources to run. But that can lead to lots of idle on-premises resources. ?Thats where the cloud enters the picture. Enterprises can spin up a thousand nodes to perform a particular gnarly job for a few hours and then knock back the compute and network resources to next to nothing when not needed. Image: iStock
6. Beyond Analytics
With better resource management features provided by YARN, database options such as HBase, and in-memory overlay Apache Spark, Hadoop becomes an application platform. More database options and middleware will run directly on Hadoop, Forrester predicts. Image: iStock
7. More Distributions, More Competition
The large enterprise vendors such as HP, Oracle, SAP, Software AG, and Tibco will create their own Hadoop distribution, as IBM and Pivotal already have. Specialized public and private cloud distributions also will emerged, generating plenty of competition and customer choice. Image: iStock
8. Linux and Windows Bundles Coming Soon
Microsoft, Red Hat, VMware, and other operating system vendors can potentially include Hadoop and make it a configurable option within their operating systems and virtualization platforms. If a technology management professional wanted to add a node to join a Hadoop cluster they could simply turn it on through configuration, Forrester notes. This trend should sound familiar: Netscape sold Web servers at hefty prices in the 1990s. Now Windows and Linux have HTTP servers built in for free. Image: iStock
Thank You and More
Special thanks to Forrester Research for the eight trends list. Information Management edited the predictions for length. Image: iStock