Select Page

CURRENT OVERVIEW

On August 13, 2012, Symantec (in partnership with Hortonworks) announced an add-on for its Symantec Cluster File System (CFS) that turns the table on Hadoop deployment.

Until now, a Hadoop cluster required a one-to-one relationship between server and direct-attached storage. It also did not allow the use of a centralized storage system. Additionally, Hadoop was oft criticized for the perceived single point of failure (SPOF) – the name node.

The new Symantec Enterprise Solution for Hadoop distributes the Hadoop workloads across multiple servers, each connecting to a Symantec CFS via a connector. The Symantec CFS presents an advanced file system that includes data efficient services such as snapshots, compression and de-duplication.

Neuralytix Perspective

At Neuralytix, we have trademarked a number of sayings as it relates to Big Data.

The first one is:

If you’re not doing it, your competitors are!™

The second is:

Stop spending money on technology, start making money!™

For many organizations, there had been too many excuses not to implement Big Data. The most common ones include:

  • It has a SPOF, therefore it is not enterprise ready;
  • It does not conform to our enterprise standards;
  • The cost of extract-transform-load (ETL) processes are too costly or inefficient to make Big Data effective;
  • Hadoop storage does not allow the enterprise to properly backup data per corporate standards; and
  • Big Data clusters require too much extra storage.

With the new Symantec Enterprise Solution for Hadoop, in partnership with Hortonworks, these common excuses are no longer valid. Of course, many non-believers of the business value of Big Data will continue to make trivial excuses. But take note of the two trademarked sayings by which Neuralytix has built its business, ignoring Big Data will cost an enterprise business, and erode competitive advantage.

The Symantec solution also solves a major problem associated with Big Data. In order for solutions such as Hadoop to access and analyze the necessary datasets, massive data movements were required – data would have to move from transaction systems, data warehouses and traditional content repositories, then imported into the Hadoop cluster. By providing a common storage interface between OLTP, OLAP, other content repositories and Hadoop, the data no longer has to be moved to Hadoop; instead, Hadoop moves to the data. This can result in a significant time to value improvement as well as reducing the amount of infrastructure needed to support the Hadoop based analytics.

Symantec has cleverly priced the new Symantec Enterprise Solution for Hadoop as a no cost upgrade to existing Symantec CFS. This pricing schema is perfectly aligned with the open-source community environment in which Hadoop was spawned and fostered. (That said, for those who are not currently Symantec CFS, it will be necessary to make an investment in software and deployment to take advantage of these features).

While the new solution is particularly exciting, Symantec and Hortonworks are not without challenges. Both companies will face stiff competition. For Symantec, the reliance on its Symantec CFS is likely to be a barrier for some enterprises as Symantec CFS may not be a certified file system within corporate standards. For Hortonworks, given the open-source nature of the Hadoop framework, it is not going to take very long before its direct competitors will enable some, if not all, of the code contributed to their own advantage. Both companies will have to compete against upcoming hardware, converged or integrated solutions.

For now, what Symantec has developed should dramatically turn the tables on the way the IT community thinks about Hadoop. This solution gives Symantec a much needed shot in the arm for its storage software division. It creates an entree for Symantec to increase its relevancy from data management to information management. It enables Symantec to be considered in a new light – one that is focused on the proactive accelerating business value creation, rather than the reactive disaster prevention software for which it is known (i.e. its anti-virus and data backup and restoration software).

[wpdm_file id=2]