Test Your HA Architecture Before Peak Season
In this sponsored post, IR provides tips on how to ensure your high-availability UC or contact center systems keep functioning while under duress.
Depending on what industry you're in, a few critical events could put your UC or contact center system to the test each year. These could relate to the economy, weather, peak retail season, open enrollment for insurance, marketing promotions, tax season, or a multitude of others situations. Here in North America, for example, we know the devastation that hurricane season can bring, broadly impacting public safety, utilities, and insurance companies.
Having confidence that your business will be able to withstand a disaster is extremely important. Sudden outages or spikes in customer traffic will bring your business-continuity or disaster-recovery plan into play.
For example, let's say you've architected and built a high-availability (HA) system you expect to carry the full load even if one of your sites is compromised by a hurricane or some other catastrophe. The only way to be sure your systems will still deliver the intended level of service is to test them in their failover configurations ahead of time. Utilities and insurance companies need to be ready for a possible onslaught of calls should a storm cause widespread damage in one of their business regions.
The Importance of Repeat Testing
During the months of normal operations, changes can happen within your network environment but not be immediately obvious. Those changes potentially impact your ability to get 100% of peak load for which the contact center applications are designed. For example, perhaps there have been network upgrades or configuration changes required to keep the systems working. Even if these changes are part of the network environment, and not directly tied to your communications complex, they have the potential to cause the system to gradually wander out of spec.
If someone looks at a configuration setting without really knowing what it's supposed to do, for example, it could be tweaked in a way that's not appropriate for a high-traffic environment. Under normal operations this may cause no impact, as the system typically runs at 20% to 30% of capacity. However, the periodic high-volume events you've planned for might push traffic levels in your system five to 10 times greater than where they're running during the rest of the year. Will your system actually function as intended when traffic levels increase to near-maximum capacity?
Stress Test Your Systems
Because delivering an excellent customer experience at all times (including to internal customers!) is critical, many of our clients engage us a month or two before peak season to generate a large amount of traffic into their solutions. Doing so provides them opportunities to verify that their systems scale effectively and can handle the increased load. Conducting stress tests provides them time to address any changes or issues that come up and gives them confidence that their systems are ready for peak traffic and will work as intended.
Companies that have hot standbys in place have an imperative to make sure those systems are actually capable of handling the traffic load. Running a load test into a hot-standby solution as part of peak season prep is a great way to make sure that the system hasn't been compromised over the last year. Any patches or application upgrades that have been applied to the production environment could have affected its ability to handle high levels of traffic.
Another key element of maintaining hot-standby systems is knowing that they can be accessed from the outside-in. Hot standbys are meant to kick into place without any user intervention. Whether the hot standbys are part of your Web environment or voice environment, IR can offer HeartBeat availability and experience testing against those systems. We'll periodically access them by dialing their secret phone numbers or unpublished URLs or IP addresses. Then we'll make sure that they're actually connected to the backend data processing systems to perform as expected. Should an emergency situation arise, anybody who is redirected to the standby system will have the information that's needed.
If you've made disaster-recovery plans that include rolling a cold-standby contact center in a semi into place, don't overlook validating that system's ability to actually take calls at peak rates. Your disaster-recovery plans probably include running comprehensive drills to ensure everyone on your staff knows what to do, but making sure your backup technology knows what to do is just as important!
The U.S. storm season has started exceptionally early, and whether you are preparing for that or for a different anticipated high-load event, you'll want to be certain that your system is ready for the increased traffic.
For more strategies on testing and preparing for your peak season, join our April 21 webinar, "Ensure your Contact Center Weathers the Storm."