How Plum Voice IVR Systems Guarantee Uptime
November 10, 2010
By Juliana Kenny, TMCnet Managing Editor
Built to handle high call volumes, and deal quickly and resolutely with disaster recovery, Plum Voice’s IVR systems possess an innovative way of ensuring redundancy. Typical IVR servers have a 1 in 10 chance of failing in the first year of deployment without redundancy systems.
With a stat such as that – 10 percent of hardware failing within the first year of operation – Plum architects fault-tolerant premise-based IVR solutions and advises its customers about how to build truly redundant telephony infrastructure.
To insure the continuous operation of IVR deployments Plum creates IVR infrastructure that is “robust enough to guarantee 99.9999 percent uptime,” according to company officials. There are certain important steps to take to achieve redundancy when deploying IVR systems at a customer’s site.
Multiple telecommunication circuits must be added from the carrier. Each circuit can stay up all year with a 90 percent success rate. Additional circuits components drop the probability of a telecommunications failure to 0.008 percent. Plum officials noted, “IVR systems can always be scaled linearly, so if you need additional capacity, it is easy to just buy another server and provision additional bandwidth. Adding a redundant IVR server increases capacity, and dramatically increases fault tolerance for the system as a whole.”
To eliminate points of exposure across application and database servers, users can build out database servers and infrastructures from scratch. Adding significant idle capacity, and ensuring a dedicated infrastructure, this development of database servers also requires the building of load balancers to distribute traffic. The more carefully built the replication system, the less failure.
The last step is to remove the final site-wide single point of failure. More than one carrier is required to receive call traffic, and Plum recommends a multi carrier set up to its onsite system clients. Plum officials note, “By adding even a second carrier, you have eliminated the final single point of failure from infrastructure. A failure of such infrastructure would occur once every 1,700 years.” Sounds good to us.
Juliana Kenny graduated from the University of Connecticut with a double degree in English and French. After managing a small company for two years, she joined TMC (News - Alert) as a Web Editor for TMCnet. Juliana currently focuses on the call center and CRM industries, but she also writes about cloud telephony and network gear including softswitches.
Edited by Juliana Kenny