Disaster Planning

Few actually plan for disasters; even fewer plan on how to recover from a disaster. Successfully recovering from a big problem requires a set of practices and procedures that are followed to help recover office computing ability after a number of foreseeable problems. These problems include a number of possibilities ranging from a power outage to a flood or structural failure of a building or to even larger scale events.

One of the most important items to have in an office is an uninterruptible power supply (also called a UPS) for your servers. This device can keep the servers running when a power outage occurs, plus it can instruct the servers to shut down in the event of a prolonged power outage, and can provide the servers with power for doing this. A UPS also protects the servers from power surges and under voltage circumstances. Being able to ignore power related problems, and being able to perform a programmed shutdown of the servers will prevent many disasters.

The ingredients for nearly all disaster recovery schemes consist of the same basic elements. These are to maintain duplicate sets of hardware and software and regularly implemented schemes for backing up key software. For many groups, rotating off-site backups of company data across town is sufficient. For other groups , duplicate data sets may be maintained in nearby or distant locations.

A very good if often over looked idea is to have a reservation with a local contractor to have a large enough power generator delivered, should you need it. We’ve directed people to PACO where for a small fee they will put you on a reservation list that will get first priority in the event of a power outage.

What many businesses do is to keep one or more extra servers on hand. An extra server can keep regularly updated duplicates of all company data. In the event of a failure of the primary server, this backup server can be put into service for most duties in short order. This is a very cost efficient solution for most the most common kinds of problems.

A much more elaborate incarnation of having a spare server is called a cluster server. As the name suggests a cluster server is a group of servers that manage a common set of data. In the event that one of the servers in the cluster goes down, one or more other servers can keep the enterprise working. Cluster servers are a key ingredient in what is known as fail-over or high availability solutions. This is the kind of disaster recovery solution required for large scale installs of Exchange server, as example, and well as many large scale databases. Cluster servers are expensive to install and maintain. If they are needed there is no substitute.

Many companies have multiple broadband connections. This is done both to distribute the broadband load and also to permit for a fail over mechanism. If one broadband circuit goes down, most communications can be redirected to another circuit.

Other solutions employ the cloud for various purposes. These include using distant locations for hosting data backups, hosting Exchange and SQL server based applications, and even hosting virtual desktops. Cloud based solutions are becoming more powerful and useful every day. The key limitation here has been due to the bandwidth available to you and the reliability of cloud resources. Recent improvements in broadband speed have permitted up to about 50 MB of bandwidth for business class bandwidth in some areas, and this has helped increase the number of workable solutions.

Give us a call at 206-763-8874 and we can help you implement a solution that is right for you.