Lets get technical: Nine 1Gig paths (all connections active and load balancing traffic) 9Gb/s full duplex mesh providing 18Gb/s of bandwidth between core switches. (Photo credit: Wikipedia)
Reliability of a system is measured in uptime, the percentage of time throughout a given window that a site is up and fully operational. Network reliability and high uptime figures are crucial to keeping users happy. One important way reliability is enhanced is with redundancies – safeguards so that if anything fails, the client still won’t experience a problem. A strong system will always have multiple redundancies in place.
Another simple way to ensure strong performance within a network is to ensure that every piece of machinery is holding up its weight (and this actually has a redundancy component as well – see below). Whenever possible, it’s better to split work up among all of your devices rather than letting one piece perform all the work. Otherwise, you see strain on one machine as the others sit nearby, daydreaming. Server load balancing is the practice of dividing work evenly between various servers.
Basics of load-balancing & how clustering is related
When a load balancer is put into place, incoming traffic – requests for information from the servers from user web browsers – routes through the device prior to hitting the servers. That point all the traffic is reaching, where the load balancer is located, is one network address. As a load balancer receives the requests, it divides work evenly throughout a server cluster.
The reason the servers are a cluster is because a cluster is a number of computers operating in the same basic manner to achieve the same objectives. In a basic load balancing set-up, all of the servers behind the load balancer are performing the same basic function – equal work based on what comes through the load balancer.
Advantages of load balancing
In the absence of a load balancer, anyone who accesses a site is hitting the same server. That server is essentially being inundated with requests (during peak times or as the site is becoming more popular).
When an upswing in traffic occurs, people visiting the site will either experience slow page loads, or the server will start denying requests. Not only will these issues make those accessing the site frustrated, but search engines will punish the site as well.
Installing a server load balancer allows the speed at which the site functions to remain high even during times of exceptional traffic. Even if a server fails (and here’s the redundancy aspect mentioned above), it has backup. All of your resources are utilized to their utmost capacity.
Another major advantage of load balancing is that it’s a simple, easily deployable process; and it’s built for scalability. If you need to add more servers to your infrastructure, you just plug them in. The load balancer recognizes any new devices and continues to balance appropriately, with any new machines taken into account.
How more complex load balancers differ from simpler ones
Essentially, the range of load balancers can be understood in terms of how well they process different types of data. These data types can be understood in terms of the OSI (Open Systems Interconnects) model. Here are the seven layers, with the top layer representing the most sophisticated and working down toward the most basic:
- Application (Layer 7)
- Presentation (Layer 6)
- Session (Layer 5)
- Transport (Layer 4)
- Network (Layer 3)
- Data Link (Layer 2)
- Physical (Layer 1)
Every load balancer can accurately process OSI layers 2 and 3. As the devices become more sophisticated, they become capable of handling the top four layers. The reason load balancers must be more sophisticated to handle application information is that the data itself is more complex. Figuring out the degree of workload needed by a particular request requires a more intelligent load balancer, so that it does not accidentally overload one of the servers by misunderstanding incoming packets.
Other functions of load balancers
Generally speaking, load balancers use network address translation (NAT) so that the IP of the exact server being accessed is unclear to the client. This relates to the “one network address” aspect discussed above. The IP address listed with the load balancer is what appears to the client (browser asking for information) to be the server. In this way, the load balancer cloaks locations, providing a security function.
Depending how the load balancer is configured by its owner, the load balancer sends out to a server. The server processes the request and sends data back through the load balancer to the client’s browser. In other words, the load balancer is not just in charge of balancing traffic; it also properly organizes requests and responses so that nothing goes to the wrong server or client.
Conclusion
As you can see, load balancing is fairly simple, but using high-quality equipment and implementing the practice correctly are crucial to a strong infrastructure. Load balancing is one of the many safeguards we have in place for high redundancy and 100% uptime. If you are in the market for dedicated, VPS, cloud, or any other hosting solution, find out why Atlantic.net is the ideal choice.
By Kent Roberts
