You may have heard of the term Infrastructure Hosting at some point. You might’ve stopped and thought to yourself “What could that even mean?”. After all, a system’s infrastructure is its base. How could that be hosted anywhere other than on site?
When it comes to traditional infrastructure (like roads or power supplies), it’s impossible for infrastructure to be hosted anywhere other than where it will be used, obviously. But, like so many other rules, that rule doesn’t apply to computing. Infrastructure hosting (more popularly known as Infrastructure as a Service, or IaaS) is actually considered a type of cloud computing. In this, the hardware infrastructure behind an application or something like that is provided by a company, often known as a provider, and the software side of things is managed by the user.
Each user gets as many resources allocated as he/she needs to meet the demands his/her application is placing on the provider. One of the great parts of having infrastructure hosted elsewhere is that it is much easier to increase capabilities quickly. The amount of resources you can host with a certain provider is nearly unlimited.
Since the amount of resources that each user might consume changes from month to month, the resources used are metered, and the user charged based on that. The more you use, the more you pay. You can usually monitor the resource usage of your application, to make things more transparent.
If you host your own infrastructure, the costs of the infrastructure won’t go down if you end up using less resources than you provided for. If you turn out to need more resources than you planned, you’d have to allocate more resources. That means significant upfront costs in both money and time, and increased running costs afterwards. There may also be downtime during the upgrade. That’s one of the reasons for IaaS’s popularity. It keeps worries about resource usage out of the user’s mind.
However, the user doesn’t really know the exact place where his/her allocated resources are, since they keep on changing places. The most that he/she knows is where the provider’s entire resource pool is located. The provider’s pool itself could be spread among multiple centers,which could be in different cities, or even countries, making things even more abstract. (‘Where’s your application being hosted?’ ‘Somewhere in Finland. Or maybe Norway.’). That’s not really a very big problem for most businesses, though.
What can be a serious downside for many businesses is that since the infrastructure is being hosted off site, the user has to access it through the Internet. That can make it unsuitable for many purposes, like, say, a server for multiple employees to access project files from. If the business were to host that server on a local server, rather than using IaaS, the server would be accessible through the local network. That means that speeds would be much, much higher, and that the Internet bandwidth that business would need would be lower than if it had used IaaS.
When it comes to an online application for general users rather than just business employees, though, it’ll be accessed through the internet no matter what, which makes IaaS a great option for such uses.