High availability is the method of ensuring that your website application is always up and running, so it’s always available to customers. Often it refers to the amount of up time that a customer will receive with a hosting provider. Typically this may be in the region of 99.99%, or with the likes of Hyve 99.999%, which offers around about five minutes of downtime a year. As well as that, a lot of customers would factor in certain elements in the infrastructure so if they’d incorporate perhaps an HA pair of firewalls which is where they going to introduce load balancing. As well as obviously various software layer redundancy in terms of using SQL, and always on, and so on and so forth. High Availability is quite important to the fact that we can obviously mitigate against any sort of downtime. In some cases for example a medical institution or organisation, if they have data that they need all the time, High Availability ensures that that data is available to the people that need it the most. And essentially we’re trying to avoid any single point of failure, so introducing stuff like load balancers, various pairs of firewalls, and so on and so forth, will eliminate that. It’s essential to running a business. Essentially it works in the fact that we’re going to design a solution with pairs of devices, ideally when you’re talking about network, or within the underlying infrastructure where we can cater for an unexpected downtime to ensure that we are offering a mission-critical service. There’s no solution that fits all really, it all depends on customer requirements, but there are certain elements that we can introduce such as replication clustering when it comes to databases. So you can use different methods to actually achieve it. One of the methods is using load balancing. So load balancing, it actually splits the traffic between two different servers, so when a customer’s… …Customer #1 comes down to a site, they go to Server #1. Customer #2 goes down to Server #2. You can also use clustering. Clustering’s when you have two instances of a database for example, and the data can actually be bought from one or two of the databases, so if one goes down basically still be taken or retrieved from the secondary database, which is again essential if you’ve got loads of transactions going through to one database. If that goes down you need to have one other one that’s straightaway available to customers or for application systems. You can also use replication. Replication is a technique of transferring data from one, so it could be a file system or a database, and transferring exactly across into another server or database. So if you had two databases, you want them to be exactly the same, so replication would mean transferring data from one, live all the time, to make sure the data is live on both. And essentially avoiding any downtime where possible. Essentially all your data is not in one place. So if one of your databases or file systems was to go down, if worse comes to worse, you can retrieve from a backup. By having a backup strategy in place, we can obviously push backups to a separate SAN for example, or even all sites to a separate data centre, therefore ensuring your data’s obviously not at risk within production. Should you want to discuss High Availability and any other solutions that we may have in store, visit our website or give us a call.