Most IT organizations, regardless of size, are under constant pressure to reduce the cost of service delivery, especially in a down market. Some organizations have gone as far as placing disaster recovery (DR) spending in the category of unnecessary expenses, with chief financial officers believing that reduced profits mean all costs must be cut proportionally, including DR. Neglecting DR can have detrimental consequences for organizations, making them less likely to bounce back when a bailout is needed following a disaster.
Excessive cuts may severely impact a company’s ability to test and improve its disaster recovery plan. Many organizations today have a mixture of both internal and outsourced DR solutions. Some businesses may be tempted to let disaster recovery contracts expire and to postpone or cancel testing activities. But is this really an effective cost-saving measure? Paying your commercial hot-site provider every month but halting testing makes no sense. These actions may save money in the short term, but they will only heighten a company’s vulnerabilities—and its costs—over time.
Common DR Shortcuts to Avoid
Making cuts to disaster recovery tempts many companies when budgets are tight. Here are some of the most common shortcuts—and their consequences.
1. Postponing disaster recovery testing
This might be seen as a quick way to save money if travel budgets and human resources are tight.
Consequences: Last year’s test results won’t suffice, and the auditor will not show leniency because of an economic downturn. Nothing could be more dangerous for your organization and your career.
Recommendations: Always test your backups and failover capabilities to ensure your ability to recover the infrastructure and business applications to meet recovery time objectives (RTO) and recover your data to a known, useable, and consistent state to validate recovery point objectives (RPO). A backup restore or a failover is not a solution if it has never been tested for recovery.
2. Delay buying new tape
Yes, almost 50 percent of the IBM i community still recovers from tape backups. Sometimes the delayed purchase is intentional for cost-reduction purposes. More often, however, someone intended to change the backup tapes but simply never did.
Consequences: Overusing tape will affect the reliability of the tape contents.
Recommendations: For those still relying on tape backups, follow manufacturer recommendations and always make sure your tapes are in good working order. Tape validation should be performed with an active disaster recovery test to ensure that the backup is complete, the tape media is readable, and the contents meet the requirements for full server restoration.
For those ready for a change, consider switching to a virtual tape library (VTL) instead. It’s tough to be 100 percent confident your data is secure if you’re using offsite storage for hire. VTLs offer improved security—primarily because you eliminate physical tape handling—as well as some extra safety measures like duplication and additional logical drives to split up backups for added flexibility. They also can help you cut down on cost as you won’t need to spend money on offsite storage anymore. Perhaps best of all, you can run backups and volume categorizing without unloading a tape for automated backups.
3. Scrimping on offsite storage costs
Daily pickup costs $40 to $60 per day plus a fee for the quantity of tapes in rotation. Budget officers might be tempted to switch to weekly pickup.
Consequences: Imagine having to tell your CEO that an entire week’s worth of data for company sales and related business activities has been lost because you tried to save $40 for a new tape media cartridge or $400 in offsite pickup charges. Your RPO just went from 24 hours to seven days!
Recommendations: You may be holding many of your archived data tapes far too long in the forgotten storage container offsite. Look for your savings there. Go visit your offsite storage location.
4. Reducing recovery capacity to save on monthly hot-site fees
Consequences: Saving pennies will cost you many dollars later. How do you plan to recover your servers? Only recovery servers are able to handle the workload necessary to run near-normal operations. Reducing capacity is a sure recipe for a second ordeal after your initial disaster.
Recommendations: You cannot randomly set up systems that are 50 percent of your current deployment and hope they will meet your current service levels. When testing your recovery, execute sample test scripts to benchmark performance and ensure the infrastructure will perform to meet your business needs when you need it most.
Better yet, consider your desired recovery time objective (RTO). Your recovery time objective is your best indicator for whether tape, VTL, real-time replication, or a combination of these is the best answer for recovering your business data. Adding data replication as part of your DR plan gives you the option to send your data in near-real-time from one data center to the next. If you can’t afford to be down for more than eight hours, you need to consider high availability (HA) technology for replication.
Finding a Balance
A more effective way to manage IT spending and expectations is to ensure that spending cuts provide a balance between reducing costs and meeting your recovery time and point objectives. An imbalance may compromise your ability to respond to a crisis, so it is increasingly vital to give disaster recovery the proper weight when making difficult budgetary decisions.
Ask yourself the following questions:
- Does it make sense to change my disaster recovery planning expectations for the business?
- Can the business concede extended recovery point or recovery time objectives?
- Do I outsource recovery testing or do it myself? Can we afford not to test?
- How do I assess the risks for my organization? Is the business aligned with IT objectives?
- How do I make the right technology choices for disaster recovery?
- Should I introduce new technology methodologies to recover critical business operations?
IT departments increasingly need to demonstrate their value both internally and externally. Disaster recovery is essential for ensuring business continuity, and a comprehensive plan for disaster recovery is more critical than ever. Saving pennies and risking survival dollars in a disaster does not make financial sense.
Recovery Without Disaster
Downtime comes in many forms and it doesn’t take a full-on disaster to destroy your data. This guide shows you what you need in order to build a strong recovery strategy that your business can really rely on.