When an organization starts its baby steps towards setting up a computer center, it’s normally a few servers hooked up on a local network. Users’ computers are connected to the same network to access enterprise applications running on those servers. To gain external access to the internet, devices such as routers or modems are put in place. For added security, a firewall is inserted between router and the local network to weed out viruses and other unwelcome bits of data from the web.
As the organization’s IT needs increase, the server farm continues to grow until it reaches a capacity whereby it is no longer feasible to store data in each server’s local hard disk. The more efficient method is to set up a SAN (Storage Area Network) and relocate all data into it. Although it sounds like a splendid idea, there are some factors to look into before proceeding ahead. A SAN is basically a collection of storage disks which run on its own network to ensure super fast transmission speeds. It’s important not to allow traffic from the server farm or local network to mix with traffic generated in the SAN as it causes detrimental effects to both and can bring everything to a halt.
To ensure fast response times when sending and receiving data, the SAN runs on its own set of protocols and is connected via fiber channels for maximum bandwidth. Although standard Ethernet connections are cheaper, they are not able to handle fast speeds as required by the SAN. Since the SAN (Storage Area Network) is one giant storage box comprising many pieces of hard disks, it requires its own management system to handle its operations. No one server controls the SAN as its disks are allocated to all servers. Via a central console, the SAN administrator switches from disk to disk as though moving around all servers’ storage disk. As added security, local users registered to one server cannot access the disks of another server.
In the event one of the disks in the SAN is faulty, depending on the RAID technology practiced, the SAN is able to rebuild data from the faulty disk onto a spare disk. Then it’s just a simple matter of popping out the faulty disk, replacing a good one into place and formatting it to readiness. If the SAN is set up well, it plays an important role in getting the organization back onto its feet during disaster recovery processes. Backup routines over SAN are also more efficient as compared to server by server backup.
Before deciding to move all data into a SAN, it’s recommended to check that the enterprise applications are able to support SAN. There are some applications which require their data to be locally stored and may pose a challenge when trying to implement SAN.
Dennis enjoys writing on wide range of topics such as SAN (Storage Area Network) and NAS (Network Attached Storage). You may visit for more details.