Disaster planning is a critical
IT responsibility
In a 2007 poll by The Economist Intelligence Unit, 79 percent of companies said that loss of access to their data for one week would put them at serious risk for failure. Despite this, many companies don’t have a tried-and-true disaster preparation plan.
Preparation for business continuity requires that you create an environment in which applications, equipment and data are readily available after an interruption within desired recovery-time objectives (RTOs) and recovery-point objectives (RPOs). To begin your continuity and recovery planning, conduct a thorough analysis of your business requirements.
Prioritize the data you need access to with consideration to RTO, RPO and total cost of ownership (TCO). A customer database, for instance, might require instant failover (the ability to transparently shift operations from a failed server to a backup server), while you might determine that a four-hour recovery is acceptable for your e-mail system. Depending on the level of importance data has to your core operations, you have several options in how you back it up and retrieve it.
Clusters provide continuity and scalability
Clustering is the term used to describe an arrangement in which servers are combined to boost computing power and overall system efficiency. Clustering allows each server to operate independently while connecting to a common storage device, such as a network-attached storage (NAS) server.
A clustered environment provides high availability since there are at least two servers in your configuration. If one server goes down, the clustering software you’re running seamlessly transfers the load from the downed server to the other server so you can maintain operations. And as you add more servers, you get the benefit of additional performance and availability. You also have investment protection since you don’t have to design your initial server configuration for peak capacity, which could leave a lot of computing power unutilized for a long time.
Mirroring speeds recovery
Server mirroring is also an effective approach to database availability and recovery. Implemented on a per-database basis, mirroring relies on duplicate servers within the same network and typically works in conjunction with load-balancing software to manage database resources. Typically, the mirrored server is located off-site to lessen the likelihood of simultaneous damage if, for instance, there is a fire or flood in your office. While one server handles transactions and processes tasks, another receives and stores duplicate data. As a result, when a primary server fails, the mirrored server takes over.
Mirroring provides instant failover capabilities but doubles the number of servers and amount of server space you require. It also increases processing demands that can, in turn, lead to performance degradation. Nevertheless, mirroring is widely used for managing online database systems and Web sites where there’s a critical need to keep applications and data accessible at all times.
Tape remains an affordable, dependable backup medium
When it comes to data backup, tape continues to be the most affordable medium. Tape backup offers large capacities, is often automated and its portability makes offsite storage practical.
But there is one major drawback to tape. The process of recovering data from tapes can take several hours or even days. After a server or electrical failure, the last thing you need is a lengthy data recovery period.
Continuous Data Protection offers multiple recovery points
To protect data and ensure fast recovery times, many businesses are turning to continuous data protection (CDP) or near-CDP. With CDP, the window of potential loss can be reduced from the length of your current backup interval to less than one minute.
The concept behind CDP is straightforward. You run a CDP solution either in the form of software, hardware or a combination of both. CDP solutions capture and record every change to data in real time at the block or file level, depending on your solution. When data restoration is necessary, you can restore to any point prior to the disruption.
Near-CDP is for businesses that don’t need a backup of every data change but would lose significant amounts of data when on a daily backup schedule. Near-CDP storage can be configured to automatically conduct backups at specific intervals throughout the day.
CDP has one benefit that’s not inherent in mirroring and clustering: it can weed out viruses. With clustering and mirroring, if a virus enters your network, it gets replicated onto the other servers. With CDP, you can restore to any point in your data history before a virus entered and corrupted your network.
Consider purchasing continuity services
In a perfect world, your business would have a backup site ready to use if you experienced a disaster. The reality is that most businesses cannot afford this. But by paying for a continuity service, you can go a long way toward ensuring you’ll have a fully functional IT infrastructure after a disaster.
The benefit of using a third-party continuity service is you don’t have to purchase and maintain a second set of equipment at a remote site. And depending on the level of failover you want, some service providers will deliver instantaneous failover and even provide complete management of your network. For greater dependability in the event of a disaster, experts recommend that your backup service be at least 100 miles away from your site so that both systems are never affected by the same disaster.
Choose, test and maintain your equipment
Network, hardware and software applications all play an equal role in optimizing disaster recovery and business continuity. Assembling the right components means paying attention to your business needs and the features and performance you require while staying within your budget. In addition, you need to take some time each month to test and maintain your equipment. Your CDW account manager can help you find the solution that meets your time and budget requirements.
%%d_amdefault_cta%%
|
|
|
|
Three steps
to preparing
a disaster plan
1. Using a current inventory of your network resources and computer equipment, examine and rate your IT assets in terms of the impact on your business if you couldn’t access them.
2. Create a documented disaster plan that accounts for all possible scenarios and outlines the procedures each employee would follow in each. Your plan should focus on continuity (remaining operable during the disaster) and recovery (the return to a normal state after a disaster).
3. Test your plan regularly to discover and resolve any flaws.
|
|
|
|
CDW Leasing
CDW offers four different leasing terms with three end-of-lease options. Leasing specialists are available to assist you with quotes, account information or any questions that you have. To learn more, contact your CDW account manager today.
|
|
|
