Many growing small- to medium-sized businesses (SMBs) struggle with management challenges associated with direct-attached storage (DAS) and are forced to consolidate their data storage environment. For many, this means making the move to networked storage such as a storage area network (SAN) or network-attached storage (NAS). Greg Schulz, founder and senior analyst with the Storage IO Group, answers some common questions about storage consolidation in this Q&A. The interview is also available as an MP3 below.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
Storage consolidation FAQ table of contents:
>> How can you determine when to consolidate your storage?
>> What are the benefits of networked storage?
>> What are the drawbacks of networked storage?
>> How do you determine what approach to take, SAN or NAS?
Another sign is when you have "pockets" of servers that have their own pockets of storage that you are looking to manage more effectively. Simplifying data backup, simplifying sharing, provisioning, allocation, etc.
So, it's not just having everything in one place, because you could very easily be simply moving the problem. If you have islands of storage attached to each individual server and you move the storage to one physical place, all you have done is moved it. What you are looking for is that shared storage functionality, so as different servers need different amounts of storage you can allocate it.
Also, by having those shared resources, you'll be able to afford having spare drives. You'll have RAID protection, so if a drive fails, it should be business as usual and the system will continue to operate. While that isn't backup per se, it's giving you availability, and without that you would be relying on your backup system.
You can go down a couple different paths, and in some cases you can introduce more complexity and more cost. It shouldn't be that way, really. People say "go to SAN, go to NAS and life is going to be good." But it's just like virtualization. You go to a virtual server and everything should be good, but in reality you are adding an additional layer of complexity. You need to be sure that that additional complexity is offset by additional flexibility and ability to do business.
One thing to watch out for in when moving to a shared storage environment is feature/function creep. In other words, it's like going into a candy store for the first time. Your eyes light up, you see all of the functionality but start drifting away from your core requirements. If you can stay on track, you can guard against that happening. Understand what you absolutely need to have, and then build off that.
You've got block or you've got file. File is your NAS, NFS or CIFS. Block is SAN. Most people think of SAN as Fibre Channel, but iSCSI is also SAN. You also have another option -- shared SAS storage. You can set up an external SAS array and share it between multiple servers using a SAS switch.
What are you trying to accomplish? Do you need block-based access or file-based access or both? How do you know? For example, if you are running an Exchange database, that would lean toward block, meaning FC, iSCSI, or shared SAS.
On the flip side, if you are doing home directories, file sharing, file serving, etc., you are going to be looking at a file-based system. Look at your applications. What are your business demands? What are you going to be using the storage for? You may find that you are looking for a shared, multiprotocol storage system -- a system that can support file- and block-based storage.