In typical enterprise WiFi (define) deployments, there are access points that provide
access and controllers that control access and management.
One startup is trying to get rid of the controller to save customers some important dollars.
Aerohive is hawking a new approach where policy and management control is distributed in a cooperative way across
access points in a hive architecture — sans controller.
“It enables a next-generation, controller-less wireless LAN architecture which gives users all the management and security they buy controllers for but without the problems associated with controllers like cost, performance and complexity,” Aerohive CEO David Flynn told internetnews.com.
Aerohive’s cooperative control access point is called the HiveAP 20 ag and runs HiveOS on top of standard hardware and a standard Linux kernel. . Flynn explained that HiveOS is based on Linux, with Aerohive sprinkling in some features for good measure.
Aerohive also uses what Flynn described as a “fairly standard enterprise
class access point,” with hardware that includes a chipset and radio from
vendors that other wireless vendors use.
“The hardware is not the secret sauce here. The magic is the software that
makes it all work,” Flynn said.
This “magic software” enables multiple HiveAPs to work together in a hive fashion to implement a controller-less wireless network. HiveAP includes stateful roaming, a feature that allows a user to move across an enterprise campus without dropping connection.
“So if clients are moving around on the network from access point to access
point, we coordinate between the access points to seamlessly hand off,” Flynn
User credentials are also handed off between HiveAPs in a roaming situation,
so users can remain logged into secure applications. The HiveAPs also
intelligently distribute network loads across the Hive to boost wireless performance.
The Best Path Forwarding feature further improves wireless
network performance by forwarding the traffic down to the highest speed
available link without the bottleneck of backhaul to the controller.
Flynn explained that in the traditional controller-based architecture, there
is a backhaul of traffic back to the controller to enable security and
“Our architecture gives the same management, mobility and security but with
a cleaner network architecture that eliminates backhaul — there is no single
point of failure and no bottleneck to a single point of aggregation,” Flynn
This article was first published on InternetNews.com. To read the full article, click here.