Thin provisioning is another technology that seems headed toward the checker status that really should not be. There are still key differences in vendor offerings. The first is how the thin provisioning allocates storage. Many thinly provisioned systems are really just dynamically allocating chunks of capacity as the volumes they are monitoring reach their allocation limits. For example if you define a 1TB volume some systems will allocate volume to its capacity rather than chunks at a time instead of a very fine grained allocation. When this chunk is filled up another chunk is allocated. What this means is that for each volume there is capacity that is allocated and not in use. For some vendors this extra allocation can be quite large, and in a storage system with hundreds if not thousands of volumes it can result in a lot of wasted capacity.
Another key development in thin provisioning, as we outline in our Thin Provisioning White paper, is the ability to migrate from hard volumes to thin volumes as well as reclaim capacity on thin provisioned volumes after data has been deleted from them. These are still relatively new features. Most storage systems do not have it yet and they could be important, especially in larger data centers.
There are other features that are often assumed to be the same yet on further examination are very different between vendors. Capabilities like deduplication, auto-tiering, compression and replication are just a few, even basic capabilities like the actual provisioning of storage can be vastly different on further review. When considering your storage options make sure you are playing chess, not checkers.
Track us on Twitter: http://twitter.com/storageswiss
Subscribe to our RSS feed.
George Crump is lead analyst of Storage Switzerland, an IT analyst firm focused on the storage and virtualization segments. Find Storage Switzerland's disclosure statement here.