Importance of High Availability: If you are reading this blog, you likely own 2-5 Wi-Fi-capable devices: laptops, mobile phones, or tablets. From employees to students, from doctors to guests, the common theme is that everyone now uses wireless as a preferred mode of access.
It is easy to understand why hospitals find Wi-Fi mission-critical to support all the devices and systems behind their complex procedures. Consider the retail distribution environment: A few moments of network downtime during Christmas could leave a shipping company unable to track packages and ultimately unable to deliver them on time. This momentary inefficiency incurs both immediate penalties as well as future losses for the firm. For schools administering exams via wireless tablets, downtime is painfully disruptive. In an outdoor environment, network downtime for a few minutes could mean losing with employees in a mine. Downtime is aggravating even in the case of my two year old, who watches the “Elmo Song” and other Sesame Street music videos on the iPad—a pause of just a few seconds will have her fussing.
The Problem: Today, in most environments where High Availability is important, the access point is configured with primary and secondary controllers. In the event of the primary controller failing, the access point is disconnected for anywhere between 10-100 seconds before rolling over to the secondary controller. This roll-over time, dependent upon the scale of APs (and clients) managed, is caused by the CAPWAP tunnels that connect the access point to controller having to be torn down from the primary controller and re-established with the secondary controller. Additionally, the client needs to re-authenticate with the AP, which in turn needs to validate credentials with the radius server. All this results in a 10-100 second failover time from the primary to secondary controller.
The Solution: A sub-second (yes, that’s less than one second) fail-over that maintains SSID connection. The latest 7.3 release (read as August-2012 – not some futuristic date) enables a second controller to be configured in “Hot Standby” mode to a designated primary controller.
The redundancy ports of these two controller appliances are connected with an Ethernet cable. In case of WiSM2, you can have a redundant blade in the same or across chassis with VSS. This connection is used to exchange the configurations, the CAPWAP states of APs, and regular keep-alives. This is how a sub-second failover can be achieved for hundreds of access points to the standby controller in case of a hardware failure or network loss for the primary Controller. This means that there is NO SSID Outage because of Access Points Stateful Switch Over (AP-SSO) from primary to the standby controller.
In addition, the standby controller also syncs the Pairwise Master Key (PMK) key cache from the active primary controller, so when the client re-associates to the access point, there is no need for the controller to re-authenticate with the RADIUS server, resulting in downtime of only a handful of seconds.
What is even better is that access point licenses on the primary controller are copied over to the standby controller. This means customers only need to purchase a low priced standby controller without purchasing duplicate licenses for managing APs!
Worry not — HealthCare, Education, Retail, Financial or any other customer segment leveraging wireless to offer increased productivity, better services, or mission-critical communication. With Cisco’s latest High Availability (HA) for the Wireless Network – you can be rest assured and always-on.
Here is a video that demonstrates the benefits of High Availability: