Shawn McCarthy | Internaut: The reasons for business continuity

When any government data center improvement is proposed, agencies are expected to justify their decisions. Yet, in tight budget times, spending on business continuity can be a true conundrum for data center managers.

For most IT projects, justifications are built around things like improved service to citizens or return on investment, often achieved by replacing expensive, outdated equipment. Business continuity is a different animal. The goal is to protect your networks and applications from the unknown. But how do you build justification around what might happen? How do you quantify the level of improvements you might need, and how much protection is too much, or not enough?
Basic system redundancy is a good place to start. When a failed system can automatically roll over to a mirrored system, continuity of operations is maintained. If your agency is struggling to justify business continuity spending, here are some pointers.

• Redundant data centers are very expensive, but individual redundant servers and applications don’t have to be. Failover systems for things like database or e-mail servers can be set up for under $60,000, including hardware and software. Such systems mirror data and allow applications to switch automatically to a backup server.
  

  
  

• System planners should show upper management potential savings for investment in redundant systems. Just compare the cost of a mirrored system to estimates of the number of people who will be unable to work if a system goes down without a redundant backup. The estimated cost of having those people idle can be a powerful point of persuasion. A 5,000-person organization could lose up to $500,000 of productivity each day a system is idle.
  

  
  

• Full redundancy is not always needed. Understand the day-to-day business needs of your agency. Agencies that don’t need real-time access to all data or messages may not need to invest in system redundancy. In such cases, traditional data backup solutions might be acceptable. But such agencies should still estimate the labor cost of the backup process.
  

  
  

• Look to companies such as the Neverfail Group, CA Inc. and MessageOne Inc. for affordable, easy-to-implement server redundancy solutions. As a system moves toward full virtualization across multiple applications and servers, consider virtualization designs such as those from VMWare Inc., IBM Corp. or Hewlett-Packard Co.
  

Government agencies often talk about system security, and spending is often targeted at hardening firewalls and developing access control systems. These are important, but application redundancy is what will keep an operation running. Too often, this part of the security mix is not considered in an overall security plan.

Former GCN writer Shawn P. McCarthy is senior analyst and program manager for IDC Government Insights of McLean, Va. E-mail him at smccarthy@idc.com.