The reality, of course, is that both have their own unique advantages and disadvantages. As we’ve discussed in a previous blog, the cloud brings obvious benefits in terms of flexibility and scalability - but at the same time, many customers insist on the dedicated server route for the security and compliance assurances of a single-tenant environment.
However, one area where things can be a little more complex is performance and availability. No-one wants to put themselves at risk of unexpected performance loss or downtime, and there are often conflicting views as to whether dedicated or cloud server hosting offers the higher level of reliability and resilience.
So, if you want to minimise unplanned downtime, which should you choose? Here’s what we think.
Redundancy benefits of cloud server hosting
The first thing to know is that most cloud server hosting providers will build some degree of redundancy into their platforms. This helps prevent downtime as a result of hardware failures, which is tougher to protect against when you rely on a single physical server.
Typically, the cloud platform will use resources from a number of different physical servers, and then if one server fails, any affected virtual machines will restart on another host with minimal disruption. In the VMware world, this is called High Availability. In the same way, storage will normally be offloaded to a storage area network - or SAN - that also has no single point of failure.
By comparison, there’s less you can do to build resilience into a single dedicated server. Some providers offer RAID and dual PSUs to protect against hard drive and power supply failures, but you’re still basically at the mercy of a single device - and of course, these measures all add extra costs to your monthly outgoings.
Cloud server hosting from TeleData uses High Availability, full VM backups and one of the UK's most resilient data centres to ensure continuity of service for all our customers. Configure your cloud server with us today.
Other performance and availability considerations
That said, there are a few other factors to consider when it comes to the relative performance and availability of dedicated and cloud server hosting.
For one thing, it’s worth bearing in mind that dedicated server customers have scope to increase resilience simply by renting a secondary server for failover. Some providers also offer hot spares as part of their service - although if you normally rent a high-spec server, you can expect the spare to be a less powerful machine and, therefore, mean a small dip in performance.
There's also the consideration in this type of scenario of application and data management. This is where a good provider will be able to add value by helping you to devise a strategy that ensures your standby machine will be ready to take the strain as soon as possible should it be called into action. How regularly will you schedule backups and synchronisation between the two machines for example, and what will this mean for your recovery time?
Finally, it bears repeating that not all clouds are created equal, and this applies to single-server hosting as much as IaaS. Just as you’d look at the quality of the hardware in a dedicated server, a prospective cloud server customer should investigate the quality and track record of their provider’s cloud platform. After all, virtual machines can crash, an overstretched cloud environment can slow to a crawl, and so on - so it’s important to get a feel for reliability and resilience before you part with your cash.