informa
Commentary

Does Every Data Center Need Storage?

As a business grows it reaches a size where it needs servers for certain functions; an email server, an application server for business financials and maybe a collaboration server to track and maintain documents.
As a business grows it reaches a size where it needs servers for certain functions; an email server, an application server for business financials and maybe a collaboration server to track and maintain documents.Many of these businesses will look to shared storage to host the data that these systems need. Larger businesses will have branch offices that need local application processing of similar applications and may assume the need for some form of local storage. Do these situations call for local storage?

Local storage shared or not brings with it additional responsibilities, space, power, cooling, data protection, dealing with expansion. As we discussed in our article "Getting Your Arms Around The Cloud" there are a growing number of cloud based storage solutions targeted directly at reducing the need for primary storage in the data center. In our testing most of these solutions will work by leveraging a hybrid storage model where the active part of the dataset is kept locally for speed of access but all of the dataset is stored at the remote storage provider's location. The local device typically either runs as a virtual machine within a virtual infrastructure or is a purpose built appliance. In either case the storage on the local system, since it is much smaller in size, can be very high speed, even solid state disk.

The value of cloud storage as primary storage is that the responsibility to protect, power and expand the organization's data is now in the hands of the storage provider not the local IT staff. Of course this means that you will want to throughly investigate the provider, making sure that they can and are performing those tasks to your satisfaction.

Cloud storage is not the only solution to reduce the amount of local data in data centers. Another good example in the remote data center case is to use WAN appliances that again act as caches to the servers in those locations. The devices now can deliver very high speed performance across standard WAN links by leveraging technologies like deduplication and compression as well as a small amount of local disk cache storage. They will also optimize the protocols that are transmitted across the WAN to make them efficient.

The value of this type of storage reduction to the larger organization is that all data is now centrally located in the primary data center making data protection and other storage management functions significantly easier. Most often these solutions are focused on file services type of traffic (CIFS and NFS) as well as email traffic (IMAP). Look for these WAN appliances to add the ability to support block protocols across the WAN as well. This allows remote servers to think they have local storage in similar fashion to the cloud storage example above.

Businesses of all sizes now have the ability to reduce the amount of storage they have to manage. WAN optimization engines can be leveraged to remove the amount of remote storage that has to be dealt with and cloud storage can reduce the amount of storage that has to be managed in the data center itself. In each case the infrastructure that connects the business to its remote locations or to the internet becomes more important. An investment in additional speed and redundancy should be considered a top priority.

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.

Recommended Reading: