Storage Area Network (SAN): In a network, various devices and resources are shared. We have the Servers, which are not directly accessed by the end-user. These Servers provide the computational power, processing and other resources to the end-user terminal which is typically a desktop computer. Now, these multiple Servers may have their own storage devices or they may share a pool of storage devices. Thus, a Storage Area Network (SAN) is a network that interconnects and provides a shared pool of storage devices to multiple servers. A SAN can also be a sub-network within the network.
A SAN reorganises storage resources in an into an independent and high-performance network. Thus, each server in the network/sub-network is enabled to access the shared storage as if the storage drive or device was directly attached to that particular server. SAN storage resources or devices may include tape libraries, hard disk drives, solid-state device, optical storage etc.
A server, traditionally, has been limited by the number of storage devices attached to it. Instead of this, a SAN gives the flexibility to network one server or many heterogeneous servers to share a common storage utility. Also, the servers may be located far from storage utility they use.
A SAN includes a variety of storage devices, servers, switches. Routers, bridges and gateways are also required to connect and extend SAN to other parts of the Data Centre Network. Depending on the budget, scale and need of the organisation, a SAN’s topology would be determined.
By utilising a SAN some management task could be simplified. This allows organisations to hire fewer IT workers or free up some workers who can then be allocated other tasks. Hence, it saves and increases man power use efficiency.
SAN may be utilised to boot servers at a remote location and hence it reduces the time and hassles which are involved in replacing a server.
(This diagram shows a tiered overview of a SAN. Here, multiple servers are connected to multiple storage systems).
A SAN facilitates data transfers in three ways:
- Server to Storage: This traditional method gives the advantage as the same storage device may be accessed serially or concurrently by multiple servers.
- Server to Server: SAN is useful for high volume and speeds communications between the servers.
- Storage to Storage: In this case, data moves without server intervention and hence frees up server processor cycle for other activities. For example, the backup of a disk device may be created on a tape device without any server intervention.
The functions or uses of a SAN(Storage Area Network):
- Provides a high speed network of devices and connecting storage devices with servers.
- They can be accessed by an application on networked servers.
- They are useful in backup and disaster recovery. Backup of data becomes more reliable and credible as the problems of different versions of the same data as it may exist on various servers are eliminated. Also, the remote copy of the data folder is enabled for disaster recovery and against malicious attacks.
- By the use of networking protocols, SAN may provide services to servers in a larger geographical area.
- Offers Simplified Centralised Management. As there is a single image on storage media hence its management, transfer and backup are simplified.
- Storage processing is offloaded from servers to a SAN and hence application performance of servers becomes high.
Organisations nowadays choose a SAN (Storage Area Network) which offers better availability, performance and flexibility than Direct Attached Storage (DAS). A DAS is simply a digital storage which may be attached to a server or computer, for example, the internal hard disk drive in a desktop or laptop. As storage is removed from servers and is consolidated at a place in the case of SAN, storage utilisation is thus improved. This, in turn, saves money and less space is required in the data centres where servers are placed. Since Fibre Channel is the architecture on which most of the SAN implementations are built, they provide better performance as compared to a DAS.
Before SANs were widely used, organisations used DAS. Smaller organisations still continue to use DAS as it offers lower costs compared to their economies of scale. But for larger organisations, the benefit of using SAN is much more than the costs involved in using it.
Various SAN Technologies
Fibre Channel is currently the most widely used communication protocol for SANs, but it is not exclusive. SAN networks rely on iSCSI communication, a mapping of SCSI protocol over TCP/IP; they also use ATA (Advanced Technology Attachment) over Ethernet (AoE), Fibre Channel over Ethernet (FCoE), ESCON over Fibre Channel, HyperSCSI (Hyper-Small Computer System interface) and some other protocols.
Download What is Storage Area Network(SAN)? in pdf – Click here