Everything You Wanted To Know About Storage Tiering And More

With its research indicating that tiering ranks in the top 10 planned storage investments for 2011, the Enterprise Strategy Group (ESG) has published its market landscape report. The report analyzes the various options, implementations, impacts and market segmentation for tiering solutions, and is among the first such analyses done, says Mark Peters, senior analyst and author of the report.

August 18, 2011

3 Min Read
Network Computing logo

With its research indicating that tiering ranks in the top 10 planned storage investments for 2011, the Enterprise Strategy Group (ESG) has published its market landscape report, Storage Tiering: The Case for, Types of, and Purchasing Considerations. The report analyzes the various options, implementations, impacts and market segmentation for tiering solutions, and is among the first such analyses done, says Mark Peters, senior analyst and author of the report.

Storage tiering, the act of placing different data on various types of storage media and systems, is a hierarchy dependent on many factors, states Peters. However, the key, he says, is the value a user places on particular data, and, in the end, all storage comes down to economics.

Tiering is not an end in itself; it is a means to itself, he says. "There is no point doing tiering without knowing what you intend to achieve."

Factors affecting where data is stored include the importance of the data in terms of performance, retention, or business value; the differing levels of service needed from users, customers, applications or business units; and the availability of capacity and management resources and funding available. The report includes analysis of 23 vendors, although that is only a subset of the total market, notes Peters. The report breaks down the market structure into five main segments: array-based migration, array-based caching, file system-based, software tools and archive emphasis.

At the end of the day, all tiering is all about economics, says Peters. "If storage was free, you wouldn't have tiering." But storage isn't free--it can be very expensive, especially at the high end, so you need to make decisions. "At the end of the day, it's about money."This leads to two conclusions, he adds. First, tiering is not just something you do at the high end, but can be looked at all the way down the stack. Second, users should think of data in terms of two hierarchies--I/O and long-term persistence. "It's not just about what goes on solid state, but something you do all through the hierarchy because it's about money."

Peters says that, even within the group he analyzed, there is a wide range of different solutions, from array-based migration tools like EMC's FAST to caching and tools. "The fact this can be done around files, pure software, or put in storage and caching, surprised me."

The value proposition is clear, though, he says, with the case studies and real-world results supplied by the vendors enough "to convince the skeptic or the procrastinator of the value tiering can bring." Results spanned from running applications four to 10 times faster, cutting storage costs by 71% and reducing storage administration time by half.

According to an EMC-sponsored IDC study released at the end of June, the world's data is doubling every two years, with 1.8 zettabytes (1.8 trillion gigabytes) to be created and replicated in 2011. Enterprise investment in this digital universe has increased 50%, to $4 trillion since 2005.

With this ongoing explosive growth in data, Peters sees no end in sight to the need for tiering. "The ultimate decision will be about economics. Tiering will disappear when we can get close to zero for storage."

See more on this topic by subscribing to Network Computing Pro Reports Research: 2011 State of Storage (subscription required).

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like


More Insights