Scaling a web deployment beyond a single server inevitably involves sharing files between multiple servers. Traditionally, this might have been handled using NFS, or perhaps Samba - but this was often far from ideal from a performance and/or high availability viewpoint.
For several years now, we’ve used a replicated network filesystem called GlusterFS to build highly available and replicated shared storage services for our fully managed customers (although more recently, using our soon-to-be-announced Object Storage service :)
Over the years, GlusterFS has expanded it’s scope and can do lots of clever things to grow to very large scale, but it’s still useful for simple high availability.
To get you started, we’ve written a step-by-step guide which should get you up and running with a GlusterFS replicated filesystem without much fuss.
If you want to follow along but you’re not already a customer, you can get signed up in 2 minutes - we’ll even apply an automatic £50 credit to get you started.
We can also help you build and manage clusters like this as part of our hands-on support or fully managed services. Drop us a line for more details.
