STORAGE-SAN, NAS, DAS: Explained
A storage solution is an important consideration for any business, especially in this age of information, AI, and big data.
An organization’s choice of a storage system can be a simple or complicated decision depending on the size and complexity of the organization’s and the extent to which they use and value their data.
Such delicate decisions and balancing acts include costs, current and projected storage capacity needs, and scalability, among others.
There are three primary options for an organization’s or business’s data storage systems, i.e., SAN (Storage Area Network), NAS (Network Attached Storage), and DAS (Direct Attached Storage).
SAN (Storage Area Network)
This is a type of storage network that allows an organization to connect storage devices like disk arrays, tape libraries, and switches to servers and access these pooled resources from any server on the network.
While previously only a limited number of storage devices could connect to a server, a SAN allows the flexibility to connect to one or more servers, and multiple users can share a pool of storage resources.
SAN is often used as dedicated storage for professional and enterprise environments.
Advantages of SAN
- Increased storage capacity
With externalized storage, storage capacity can easily be increased to the network’s maximum limit. A SAN can support hundreds of drives using disk arrays.
A disk array is a group of disks that work together to achieve higher data transfer and input/output rates than those provided by single large disks.
- Better performance
As opposed to traditional file server and storage solutions, SANs, using block storage, connect the data source directly to the target, which increases throughput and delivers faster performance.
- Enhanced availability
Since storage exists independently in a SAN, it’s more fault-tolerant than a server with disk arrays connected to it, for instance.
The malfunctioning of one or two disks will not have adverse effects as the other functioning disks will continue working unencumbered, making it a reliable storage solution for mission-critical stored data and disaster recovery.
When to use a SAN
- For transaction processing systems, e.g., credit card processing and telecommunications, which require a large amount of throughput
- For enterprise applications that need to be centrally managed and facilitate disaster recovery and thus need to be on a shared platform
- For database servers used in online transaction processing (OLTP) and online analytical processing (OLAP), high-traffic e-commerce sites, and such that require high-performance block-level input/output that can handle thousands of transactions per second.
- For virtualization hosts, virtualizing servers utilize SAN because of its easy scalability and its capacity to deliver high performance.
NAS (Network Attached Storage)
A NAS device can provide file-level access across the network, using NAS storage protocols such as NFS, to multiple devices and user groups.
Windows servers using the SMB protocol to share storage via the network can also be regarded as NAS.
NAS can be scaled up by adding more drives to the node or using disk arrays, and scaled out by adding performance capabilities in addition to storage capacity through adding NAS appliance nodes.
Advantages of NAS
- Cost
NAS devices use regular Ethernet networks rather than the costly fiber-channel network used in SAN.
- Flexible storage capacity
NAS devices can be expanded by simply adding more nodes or drives as needed. A NAS can start at a small scale with a few terabytes of storage capacity and be scaled up to a petabyte scale by adding more nodes.
- Simple management
NAS devices are easily manageable as they use regular or standard protocols and file systems that most users are familiar with.
When to use NAS
- Home networks: for remote employees who need to consolidate their business and work files on a secure device that can be centrally managed
- Departmental servers: for workgroups or departments in an organization that need to share files and other digital assets securely on the LAN
- Remote access: for organizations with multiple offices, branches, or remote staff who need to access the same set of data at once.
- Big Data: NAS can be used to store large volumes of unstructured data, such as IoT and video surveillance data.
DAS (Direct Attached Storage)
This is a storage device that directly attaches to a PC, laptop, or server. With DAS, no network connection of any sort is used, as opposed to NAS and SAN.
Advantages of DAS
- Cost
Since it does not require or utilize a network connection, it’s the least costly of all the storage modes.
- Simplified management
It’s simple and straight-forward, with no protocols to setup, manage, or configure, and almost every user is familiar with it, so it requires no further user training.
- Flexible capacity
Since it’s a plug-and-play affair, a user can simply add another hard drive to a PC or server without the need for specialized IT personnel to help with setup and management.
When to use DAS
- DAS is a viable option for small businesses or individuals looking to expand storage capacity without paying for expensive features that they might not need.
- Workstations: when there’s a need to attach additional hard drives or external storage devices that can hold large files on individual desktops, laptops, and even security and video management systems
—
While DAS connects directly to the server, NAS and SAN use the network to create storage pools, which are accessed and managed over the network.
The main difference between SAN and NAS is that SAN uses block-level architecture, while the latter uses file-level architecture. Block storage separates metadata from the data set, while file-level storage processes the complete dataset, including its metadata, which is thus the reason why SAN systems are faster than NAS systems and are more aptly suited for heavy enterprise-grade workloads.
DAS can be cheap and relatively fast, but it also creates data islands, which increase inefficiency and the possibility of human error.
What determines the choice of storage mode differs from organization to organization. Some of the factors include:
- Total cost of ownership (TCO): depending on the unique organization, onsite storage might be more or less economical than outsourcing storage to cloud service providers.
- Storage capacity: data-intensive businesses have vastly different needs and priorities than organizations that only have to deal with a few files or gigabytes of data.
- Data security and protection: regardless of size, all organizations prioritize data security, some more so than others. The more critical data an organization handles, the greater the need for secure and robust storage. This also extends to the safety of data against unauthorized access.
- Performance: data-intensive organizations that rely on and process large amounts require higher levels of performance, which informs their choice of storage.
- Infrastructure scalability: an organization that foresees a situation where incoming data volume will increase and need to be stored for processing will prioritize storage modes that are easily scalable.
- Human resources: the number of people or staff a company is willing to employ in managing the data infrastructure
