If you’ve sat through a few meetings on storage needs, you’ve no doubt heard someone refer to NAS (Network Attached Storage) and SAN (Storage Area Networks). There’s also a good chance these terms were used interchangeably, which can cause a lot of confusion. I’ve heard some people refer to one or the other when they mean “any storage that resides on the network”. That definition works for some casual discussions when IT isn’t around to set the record straight. But it’s not helpful when your company is trying to decide which to choose. So I’ve worked on a NAS vs SAN comparison to give you an idea of how they work.
I’ll compare the two technologies and then make some recommendations based on use cases. As with any major IT purchase, it’s always wise to speak with your MSP who can help assess your specific needs. This is especially the case with a SAN which typically require specific equipment and the right skills to implement properly.
Before we get started, let’s define the two storage technologies. We’ll start simple and move up the food chain as we go along.
NAS is generally considered the simpler technology of the two. In fact, a NAS might look like a server to most people. NAS solutions typically use TCP/IP networks such as Ethernet. Employees access files on the NAS as if they were stored locally. In terms of usage, a NAS is often faster in terms of moving files, but not as efficient as a SAN.
Due to its structure, a NAS has limited scalability and are LAN dependent in terms of performance. Not only is a NAS easier to deploy to an existing network, but it’s usually less expensive as well. In short, a NAS is a single device/server/appliance sharing its own storage over a network. There’s a very good chance you have a number of NAS already in use at your business.
SAN is a more complex technology which makes it more expensive to deploy in most cases because it requires specific equipment. It usually connects to a network via Fibre Channel or iSCSI (Internet Small Computer System Interface), and users access blocks of data instead of files. In short, a SAN appears to the user as local storage, but can be managed centrally. A SAN works best in environments that require fast I/O such as database and e-commerce websites. SANs also have better redundancy and scalability and run independent of the LAN. In short, a SAN is a dedicated network of devices all working together to provide block level storage.
Before I discuss some use cases for each storage technology, I want to point out that most enterprises utilize both technologies. While there’s some overlap, most storage needs fall into one camp or the other. In most scenarios, the usage case will dictate which technology you should use. In the real world, budgets, knowledge and “use what we have” also come into play.
Use Cases – NAS
File Storage & Sharing
: Let’s start with a simple storage need of every company: file storage and file sharing. Your company has employees who create files and share them with others. This is the most popular usage case for deploying a NAS, especially for small to medium sized companies. A NAS is great for those who create typical Office documents that aren’t huge in size. Smaller companies like the flexibility of the NAS because it’s basically a server full of drives. You probably already have the components to build a NAS.
There seems to be some controversy on this topic, but plenty of people are using NAS devices from companies like QNAP
to run their VMs. Of course, you can use a product like FreeNAS
to host the VMWare data store on a NAS you build yourself too. There are literally hundreds of personal and small business focused NAS devices on the market that will work.
: Microsoft added NAS support a few years ago, and they appear to have made it a priority. Similar to VMWare, a number of vendors sell devices they’ve optimized for Hyper-V. Utilizing a NAS for your virtualization environment can make a lot of sense.
Use Cases – SAN
High-traffic sites with thousands of customers who are processing orders every few seconds will benefit from the speed and efficiency of a SAN. SANs architecture has been optimized for fast I/O which is key to keeping customers happy.
Again, the speedy I/O and fast transaction times of the SAN are what make large databases sing when handling thousands of calls a second. The bigger and more critical the DB, the more sense it makes to use a SAN.
When you have many critical servers in your data center running applications that you can’t afford to go down, then a SAN is your best choice. The SAN also allows for a quick recovery if disaster strikes. One caveat here is that a SAN probably isn’t your best solution if you need a DR plan for a single application. In that case, consider using a software solution
rather than a more expensive hardware solution that requires a SAN.
If you’ve made it this far, I hope you come away with a better understanding of NAS and SAN. I also hope you have a better idea which storage technology works best in various scenarios. There are hundreds of use cases for each technology, and I only provided a handful today. My hope is that you’ll have come away with a better understanding of each and can apply them to real-world scenarios at your business.
Keep in mind that cloud service providers also offer their own version of cloud-based NAS and SAN products. One reason for this the proliferation of mobile devices that are often not connected to a corporate network where the traditional NAS and SAN reside. Companies like Microsoft, Amazon and Rackspace and creating new products that extend the NAS and SAN to the cloud. I’m certain that we’ll see continued innovation with both of these mature products.