Around the Storage Block

Eliminate data fragmentation to ensure business data storage success


Disjointed secondary storage can cause mass data fragmentation, which can slow down the business and potentially raise compliance risks. 

HPE_Cohesity_mass data fragmentation_blog.jpgAs you look across your business data secondary storage landscape and the various services you operate, you may find yourself thinking, "Man... we have a lot of stuff out there." You may see some file services in one department, a poorly managed analytics project in another, and a series of data protection tools used across various other departments.

What you're experiencing is the sudden realization that you have a serious data fragmentation problem.

When secondary storage becomes a primary problem

What you have is an issue with your secondary storage, and you're not alone. According to research performed by VansonBourne, 87 percent of respondents say that this secondary storage fragmentation is making their jobs harder and their organizations less efficient.

What, exactly, is secondary storage? Simply put, it's everything that you don't consider to be a primary workload. Primary workloads are those that have been designated as mission-critical and have exacting performance and availability requirements upon which the business depends. Everything else is a secondary workload. That's not to say that secondary workloads are unimportant. They are still critical to the organization. They just may not have the performance requirements of those primary workloads.

Examples of secondary workloads that house critical business data storage include:

  • Backups—These may exist across one or more repositories and storage mediums.
  • Disaster recovery—You may have disaster recovery happening environment-wide or per application in some cases.
  • Files—You know you have that one file server that is the dumping ground for when no one knows where else something goes or someone is too lazy to find out.
  • Object files—Your unstructured data needs to go somewhere, and it likely lives in an object store in your data center or elsewhere.
  • Test/dev—Test/dev virtual machines and workloads are probably strewn about your environment, including many you may not even know about.
  • Archives—This is the old stuff that you know you have to keep or, in some cases, the stuff from 1996 that your company keeps, "just in case."
  • Analytics—Your team made a copy of the transactional database every day for the last three months, and they're using those copies to run analytics so they don't impact production. Or, perhaps you're simply creating separate data sets that need to be stored somewhere.

As you look at the list above and the descriptions of each, I suspect that you may have nodded your head at many of them. You will also have noticed a key feature of all of these items: they're everywhere. All of these items are massively fragmented across a complex patchwork of specialized point products and infrastructure silos in the data center and in the cloud. And that's the problem.

What you can't see can hurt you

Like a rattlesnake hiding in the brush, as you wade through the forest that is your fragmented data landscape, you might find yourself getting bitten. There are a number of opportunities for risk to raise its venomous head in a fragmented data environment:

  • Cost—Fragmented data is inefficient. You're storing all of that data in separate silos with no benefit for potential economies of scale. Further, since you have limited visibility into what you're storing, you may find yourself becoming an e-hoarder, holding onto data that the company simply doesn't need. Of course, you're still paying for that storage capacity, even if the company never actually uses the data.
  • Loss of efficiencies—Secondary storage silos are growing really fast, with respondents to the VansonBourne survey indicating that they're seeing growth rates of 25 percent, 50 percent, and 75 percent or more each year. This isn't sustainable unless you can easily apply storage reduction technologies such as deduplication and compression against it. Unfortunately, the nature of many fragmented environments means that these storage efficiencies simply aren't happening.
  • Lack of data integrity and too many data copies—Some estimates say that companies have as many as ten copies of data floating around various silos. This raises some serious questions. Which copy is current? Which copy is accurate? Which copy is complete? Now, rather than spending time gleaning insights from data, you're spending time trying to figure out which data set is the right one.
  • Compliance—Are you storing sensitive data that could land you in compliance jail if it gets out? Are you adhering to all regulations for your industry for all of your data, including those data elements that may reside in various secondary storage silos? Most important, how do you know with 100 percent confidence that you're in line with regulations? Do you really have sufficient visibility across your secondary storage silos to say for certain?
  • Opportunity cost—I hinted at this above, but inefficiencies around this secondary storage environment also make it far more difficult, if not impossible, to focus on moving the organization forward. You're mired in data quicksand, constantly looking for that rope to get thrown your way so that you can pull yourself out but, as soon as you start making some progress, some new data sinkhole opens beneath you.

Facing the challenge of mass data fragmentation

Secondary storage sprawl is a growing problem that carries high costs and risk to the organization. Attention is being diverted away from critical business needs as IT scrambles to keep up. There is hope, though. Look for the solutions that have hit the market designed to help you handily solve these problems and bring your secondary storage assets under a single umbrella. These solutions, which rely on hybrid IT and converged infrastructure, can prove to be revolutionary for organizations that take advantage of them.

Learn more about what these solutions are and how they work: Discover the data storage solution from HPE and Cohesity that collapses secondary storage silos

Scott Lowe.jpgMeet Around the Storage Block blogger Scott D. Lowe, CEO and Lead Analyst for ActualTech Media. Since 1994, Scott has helped organizations of all stripes solve critical technology challenges. He has served in a variety of technical roles, spent ten years as a CIO, and has spent another ten as a strategic IT consultant in higher education. Today, his company helps educate IT pros and decision makers and brings IT consumers together with the right enterprise IT solutions to help them propel their businesses forward.

0 Kudos
About the Author


Our team of Hewlett Packard Enterprise storage experts helps you to dive deep into relevant infrastructure topics.