New laws such as the Federal Electronic Signatures in Global and National Commerce Act (E-Sign) and the Uniform Electronic Transactions Act (UETA), which has been adopted in 38 states, raise the stakes for e-mail transactions that are lost in-flight or left undelivered. These laws allow the use of electronic documents as official, legal notices (with some exceptions) and legal contracts. They define an e-mail as being sent when it leaves the sender’s e-mail system. As a result, e-mail that is not delivered to the end user becomes, in effect, a lost contract and a potentially significant legal liability.
Independent software solution providers such as Lucent, Dialogic, Veritas, GFI Fax and Voice, Point Solutions, and others have created useful complementary products that have given Exchange a high level of growth in the enterprise communication market. These applications include e-mail security, unified messaging, wireless services, storage systems, backup utilities, system metrics, administration, and others.
Responsible IT professionals are looking for ways to minimize Exchange’s susceptibility to failure. Exchange-based applications have become so integrated into the fabric of today’s businesses that they demand the highest levels of availability and “best practices” system administration.
Exchange is generally perceived as being the best communications application for companies. So the question now is, “Can companies afford a minute of Exchange downtime?”
To help determine this, IT departments have to take a look at the types and ranges of
Exchange-based services they provide. Then they must determine how integrated into the business environment each service is. Finally, they must assess the quality of the business environment: how good are its systems, and how detailed are the procedures to administer them?
IT professionals also need to organize system assets into an inventory checklist. They can use this checklist to determine whether they have a sound systems model for maintaining e-mail service or whether the model needs improvement. The simple exercise of identifying system assets and noting their use and control will help determine preparedness for enduring a system failure.
Determine Costs Of Downtime
The next step in the process is to understand the actual costs associated with an Exchange down event.
The cost of system failure can be measured in many terms, including:
• Employee productivity
• Business transactions that cannot be made
• Labor hours and resources required to recover the system
• Damage to your company’s reputation
• Employee morale issues
Few businesses in today’s marketplace can risk a system down event. In his white paper “What Will It Cost You? Microsoft Windows Clustered Servers and Application Availability,” author Lou Finnegan of Advantage Product Marketing estimates the cost of downtime between $1,000 and $5,000 per hour, depending on the number of users affected. Clearly, e-mail server downtime is expensive.
Eliminate The Cause Of System Failures
The best way to eliminate Exchange downtime is to eliminate the cause of system failures. The more resources used to prevent a failure from occurring, the fewer hours spent in recovery. Just as in heathcare, preventive medicine is the most cost-effective method in the long run. Eliminate the chance of system failure by making the system environment as strong as possible. In the event that Exchange does go down, invoking a fast, effective recovery process is the next best strategy businesses can use to limit downtime costs.
The most common cause of downtime, as well as data loss or corruption, is hardware failure. One way to reduce occurrences of hardware failure is to adopt a “best practices” policy. This can be quite time consuming. Most IT professionals have a clear understanding of the practices needed to maintain a top-notch administration environment. The key is to employ them consistently, with constant attention to detail.
Maintain Effective, Efficient Backup-And-Recovery Strategy
A sound backup-and-recovery strategy, more than any other practice, will help reduce the chance of data loss and speed the time to restore. Remember to match the Exchange logging method with the backup strategy. (If full backups are performed nightly, then retain circular logging. If you need incremental or differential backup, then turn circular logging off.)
Schedule Hardware Maintenance
Physically clean all boards and components; defragment the disks if warranted; monitor logs to determine if there are periodic faults requiring attention. Check your capacities. Run diagnostics and test the UPS.
Plan Software Upgrades
Take preemptive action against potential system software failure by researching your need to upgrade and by reading service bulletins.
Maintain Backup/Restore Capabilities On Exchange Server
Keep a fast backup tape drive on the Exchange system itself to reduce time to restore.
Organize AdministrationPolicy With Small Team
Share the responsibility. Discuss the issues. Rotate the administration responsibility so everyone can solve potential problems. Hold administrator training regularly. Keep up with changes and new developments.
Practice Disaster Recovery
Test your methods of Exchange or disaster recovery. Recovery procedures are performed infrequently. Your ability to perform these procedures will atrophy if they are not practiced regularly. Once per month is ideal; once per quarter should be adequate.
Resist Relaxing Standards
Downtime occurs infrequently. You can be lulled into believing it won’t happen at all. Rest assured, the moment you are caught with a bogus backup tape will be the time your system fails.
All administrators understand action must be taken to prevent any critical system failure, including Exchange. Given unlimited resources, an exceptionally reliable Exchange environment is easy to attain. However, with budget and staffing constraints, IT professionals should apply the tools and techniques that will reap the greatest benefits for the smallest investment.
Those solutions that preempt downtime at its source have the greatest return on investment. To achieve the highest levels of Exchange availability, you must recognize that, no matter what you do to help mollify the impact of the failure; removing the causes of failure is the most cost-effective way of dealing with the problem.
Consider using a fault- and disaster-tolerant solution for Exchange. These solutions are a cost-effective way to maintain 99.999 percent Exchange server availability and provide complete protection of data in the event of a hardware fault, transient software failure, or disaster.
Margaret M. Kelleher is marketing manager of Marathon Technologies. With more than 16 years of experience in high tech and engineering industries, Kelleher is responsible for communicating Marathon Technologies’ key product messages through print and electronic media. More information on Marathon Technolgies at: www.marathontechnologies.com.