A New Twist on an Old Technology to Achieve Enterprise Network Storage Optimization; Not all Wine and Roses

By January 21, 2009 Data Reduction, DCIG

I’m not exactly sure how the weeks and months slip by so quickly. Over the past few months I’ve been meaning to cover a number of new technologies that have caught my eye and could really help companies get more bang for the buck out of their existing storage infrastructure investment. One such product that falls in this category is the Storwize STN-6000 Series of capacity optimization platforms. If that sounds like a mouthful, it is, but the 6000 Series offers a rather unique value proposition that no other product offers in quite the same way.

First off, what it does. The STN-6000 Series resides in the data path on corporate LANs between production servers and corporate file servers and compresses data stored on the corporate file servers. While it supports any file server that does CIFS or NFS traffic (which is pretty much all file servers) and is available in models suitable for departments, organizations that are using enterprise network filers like the EMC Celerra, HP StorageWorks 9100 or NetApp FAS6000 Series are likely going to see the greatest benefit. The simple reason is that organizations need to generate enough savings in capacity to justify the cost of introducing the 6000 Series into their environment.

Of course, anyone who works in an enterprise data center knows that introducing an appliance in front of an enterprise production storage system is not a step one takes lightly. While Storwize claims that it can achieve capacity reductions of 10 to 15-fold on these systems, enterprises and the individuals who manage them look at more than just the money such an appliance will save. The guys who manage these enterprise NAS boxes know sometimes better than the vendors selling them that if a production application goes offline or experiences problems after the appliance is introduced, the amount of money they may loose may far exceed whatever savings the appliance delivered (and their future at their company may not look so bright after such an episode either).

To try to alleviate some of these fears, the 6000 Series differs in some important respects from other in-band appliances.

  • Enterprise caliber features. Not that other in-band appliances don’t offer this functionality but one of the first things an organization is going to look at is does it meet or exceed the availability, data integrity, performance and reliability requirements that their current enterprise file servers deliver. While I can’t speak for all enterprise NAS enterprise environments, from my perspective, the 6000 Series certainly would seem to meet the requirements of many shops. It supports link aggregation across multiple 1 Gb/sec Ethernet ports, is available in clustered configurations for constant availability and provides end-to-end verification of the data for data integrity.
  • It does NOT do file or storage virtualization. This might be one of the hardest things to wrap one’s mind around when looking at the 6000 Series. Most of the time when vendors want to introduce an appliance into your environment, virtualization is always lurking in some form or another in the background. This is not the case with the 6000 Series. It compresses data that is stored on file servers using the proven and known lossless Lempel-Ziv (LZ) compression algorithm. What makes this appealing and somewhat reassuring to companies is that it creates no direct dependency to have the 6000 Series present in their environment to recover the data since the metadata about how to uncompress the file is stored along with the compressed file.
  • Requires no change to your network infrastructure. Again, this product substantially differentiates itself from virtualization solutions which often require data migrations, network reconfigurations, integration with network security protocols and outages to the environment. The 6000 Series does not. Instead it takes advantage of existing IP network protocols such that files and data to be stored on a file server first passes through the 6000 Series where it is compressed and then stored on the target file server.

The 6000 Series is really a revolutionary way to compress and store data on network file servers. In this case, what makes the 6000 Series so revolutionary is its simplistic design and ease of implementation (which almost pains me to say because I almost always end up eating those words). But the immediate upsides of reducing storage capacity, no need to make changes to your network infrastructure and no requirement to introduce virtualization with options to gradually deploy the technology and remove it if so desired would seem to make the 6000 Series a no-brainer for most enterprise network file serving environments.

Of course, the issue that enterprises will wrestle with in this economic environment is determining if Storwize is a viable long term play and can they actually count on Storwize being around in the months and years to come if these economic conditions persist? My sources tell me that Storwize let its Sales VP go earlier this month which indicates that enterprise organizations are not just backing up the truck to load up on Storwize’s product just because it is great technology. Like everyone else, Storwize needs to justify its value and in this environment, that is easier said than done.

Jerome M. Wendt

About Jerome M. Wendt

President & Founder of DCIG, LLC Jerome Wendt is the President and Founder of DCIG, LLC., an independent storage analyst and consulting firm. Mr. Wendt founded the company in November 2007.

Leave a Reply