In Pursuit of Elastic Data

Michael Vizard

As the amount of data that needs to be managed continues to grow, the IT industry as a whole is struggling with ways to cope with the problem.

Right now, most of the solutions being discussed can be broken down into three camps. First, there are the various "Big Data" offerings, such as the Hadoop technology developed by Google. Then we have the rise of distributed cache servers that process more data faster than ever in memory. And finally, we're starting to see any number of new file systems that promise greater levels of scalability than ever before.

Most IT organizations are a long way from deciding what approaches to this problem they will ultimately take to address this problem, but one thing they know for sure is that their data storage costs continue to rise. While it's true that the cost of a terabyte continues to drop, the amount of data that needs to be stored is increasing so rapidly that the total cost of storage continues to grow year over year.

This may create a situation where, between the need for a more elastic approach to managing data in both pirvate and public cloud computing scenarios and the real desire to cut costs, IT organizations are more willing than ever to try some new approaches to storage management.

One of those new approaches is an open source file system from Gluster. The basic idea is to layer an open source file system on top of industry standard Intel servers to create a network attached storage (NAS) system. But the underlying file system is a pretty advanced offering in terms of its ability to use an approach to hashing to create a elastic file system that keeps track of where every piece of data is without impacting performance. The file system essentially creates a large pool of virtual storage made up of virtual volumes. The data itself is written to disk by the underlying operating system, so no proprietary layer of software gets in the way of the applications and the underlying data.

Gluster CEO Ben Golub says this creates a truly elastic file system that doesn't need to rely on metadata managed by an object-file system that requires changes to applications, or tweaks to existing file systems that only scale so far on expensive, proprietary NAS hardware.

Golub sees the Gluster File Systems as part of a wave of open source technologies that are changing the economics of IT. Data storage remains one of the last bastions of proprietary technology that Gluster Technologies wants to breach.

Whether Gluster wins the day or some other elastic approach to managing files emerges, the fact the matter is that we're on the cusp of a major transformation in data management. The only question at this point is not whether it will happen, but rather how long will it take.



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.