Glossary

Fault tolerance

Fault tolerance is a system's ability to keep running when one of its components fails: a server, a disk, or a network link. Every critical part has a backup that takes over automatically, so customers can still get through.

Why it mattersIf your contact center goes down, customers can't get through. Fault tolerance protects you from that downtime.

Also called: Fault-tolerant system, High availability

In practice

How it's achieved

A standby server for each main server: if the main one fails, a standby with the same configuration takes over
A cluster of several servers that share the load
Mirrored disks (RAID): data is written to two disks at once, so losing one loses nothing
Two network links and two carriers: if one goes down, calls go through the other
Backup power for servers – UPS units and generators in the server room or data center

Example

How many servers you need

You add a standby server for each main server, so the server count doubles.

For example, a 20-agent contact center needs 2 main servers: one for telephony and one for everything else. With fault tolerance, that's 2 + 2 = 4 servers. At 70 agents, you need 4 main servers, or 8 with standbys.

Comparison

Fault tolerance vs. backup

Fault toleranceWork doesn't stop: the standby takes over automatically, and there's nothing to restore.

BackupYour data won't be lost, but the system has to be restored, which means hours of downtime. You still need backups with fault tolerance – to protect against errors and accidental deletion.

In CONTACT.WORK

How it works in CONTACT.WORK

For on-premises installs, a cluster provides reliability: every main server has a standby. Our server calculator shows how many servers you'll need with standbys. In the CONTACT.WORK cloud, a standby server is optional and costs the same as the main one, and we monitor it for you.