DotBlock's load balancers are designed to assist in reliability and performance of websites during high-traffic periods. By distributing requests across multiple servers, the load balancer ensures seamless user experiences by balancing out network traffic.
for your benefit
When your load balancer decrypts SSL traffic before passing the request on to the server, it is called SSL Termination. Offloading this process improves load time by freeing servers of the extra CPU cycles needed for decryption.
Within load balance groups, a server pool can provide redundancy among that collection of servers. They allow your configuration to use next-hop groups to forward duplicate packets used in port mirroring.
The purpose of a load balancer is to create a virtualized system that parallels the functionality of the physical web server, across a node network. The system adapts to increased traffic without impacting existing performance.
Numerous algorithms are utilized to sensibly distribute client requests across server pools. The specific formula for your server will depend on the type of service or application being delivered and the status of the network of servers at the time of the request.
Persistence is the concept of ensuring that a user stays on the same server once a session has begun, or when the user resumes a previously suspended session. This can be implemented based on the source or destination IP address of a packet.
Health checks are periodically performed on your server network. The load balancer sends a request to the node and will mark any unresponsive node with a "down" status. Checks are continually performed regardless of status, to ensure availability.
The load balancer can determine if a node is down and unable to receive requests from a user. Similar to health checks, the load balancer will actively monitor an unavailable node until it can be marked as available again.
Compressing responses often significantly reduces the size of transmitted data. However, since compression happens at runtime it can also add considerable processing overhead which can negatively affect performance.
In the hosting/data world, redundancy is extremely important for usability. If any one component goes down, a redundant component can step right in without any noticable difference.