Load Balancers

Load balancers are used to distribute traffic evenly across a networt, minimising response time by voiding overload and adding redundancy by spreading load across multiple nodes. Load balancing allows seamless replacement of failed nodes, efficient resource allocation and maximum use of the available capacity.

The CatN platform uses four dedicated servers to balance load across the static and dynamic content web server tiers. Each layer has load distributed by a pair of servers adding redundancy and removing a single point of failure. Due to the nature of the network tasks the load balancers provide, they must be stand alone devices.

Our load balancers act as an intermediary between user requests and the backend servers. Requests are forwarded to a server in one of our layers (static/dynamic) which replies to the load balancer which then serves the request to the user. The user is not aware of the mechanism being provided by the load balancer which has security benefits for our network by hiding our infrastructure from web requests.

To prevent users requests from being directed to the same server, unbalancing load across a layer, we have disabled keepalive and persistence on our load balancers. This prevents multiple requests on a single session being continuously directed to a single server making the spread across all the layer nodes unbalanced, having an impact on resource efficiency and the single servers performance.