Mammoth VPS customers can purchase access to a HTTP/HTTPS load
balancer for $9.95 per month. The load balancer is fully redundant
and configurable through our control panel.
Screenshot of of load balancer configuration
- At least two VPS to gain high availability
- Each VPS must be running a web server such as Apache or
- Each VPS must serve essentially the same content
How it works
After purchasing load balancer access, you will be assigned a
new, "virtual" IP address. This will be the IP address of your web
site. Using our control panel, select the VPS that HTTP requests
will be distributed over - these are referred to as the "real"
The load balancer includes an automated health check - every
three seconds, each of the real IPs will receive a HTTP request to
determine if that server is functioning correctly.
When a new HTTP request is received at the virtual IP, the load
balancer will then select a healthy real IP and pass the request
through. The real server then processes the request and sends a
- The load balancer is non-terminating: packets from each client
are passed directly to the server with no modification to the
- The load balancer beforms layer-2 ("direct") routing: each
packet from a client has the destination MAC address altered and
then placed back onto the network for the real server to
- From the real server's perspective, the source IP is still the
client - the load balancer is invisible.
- From the real server's perspective, the destination IP is the
virtual IP. For this reason the virtual IP must be added as a
secondary address on each of the real servers. Our virtualization
stack will automatically block ARP on the real servers to prevent
- The load balancer has a 60 second timeout on session stickiness
(requests from one IP address will continue to be sent to the same
real server within the 60 second window)