Chances are you are currently paying for hosting in some form, but do you fully understand what it is you are paying for? You may have been assured by your team that you need it, but not fully understand why.
In this tech translation, we help break through the technical jargon surrounding these essential services.
What is Hosting?
Hosting is a blanket term for storing data on a type of computer called a server. Servers can store a variety of things such as files, emails, or even whole websites. When you go to view any of these items, your own computer or device sends a request for the information through the internet to a server, which then sends it back. The host is the company or person who owns and maintains the server.
Hosting can vary hugely in complexity, from very simple cPanel setups to incredibly complex arrangements covering failovers and redundancies, and this can make it daunting to find the correct solution for your needs. We are going to talk in terms of website or web app hosting, as this is the most common paid-for service. These are some of the options you might be confronted with, and an explanation of what services your host actually provides;
The most common option for web hosting, shared hosting is when lots of different people’s websites are kept on the same server. This is a low-cost option, but it does come with some risks. For example if one website on the server becomes corrupted, it can spread to others on the same infrastructure. If a website also encounters a large load (lots of visitors at once - for example during a sale, or even a DoS “Denial of Service” which can happen when someone deliberately injects crappy code to cause trouble on a site), then it can slow down, or even take down, the whole server and affect other websites kept there too.
While as a client you are responsible for maintaining your own code, hosting companies will of course put measures in place to reduce these risks to the server; such as firewalls, encrypted uploads (SFTP) and continual monitoring.
Dedicated hosting is when you have a server to yourself. This will cost more, but it means no-one else has use of the same infrastructure, making it both more reliable and more secure, as it mitigates some of the risks with shared servers. This of course doesn’t guarantee 100% uptime - there’s no accounting for hardware failures or even Dedicated DoS attacks. Your host should again take measures to prevent these issues, as well as you adequately maintaining your own code.
‘Cloud’ is a term that is quite broad in its meaning. In the context of hosting, cloud usually refers to a network of physical servers, designed to work together to provide more resilient hosting, including built-in protection against the failure of one specific component (failover) and the ability to use more or less physical hardware depending on need (scalability).
Cloud hosting makes use of virtual servers. A virtual server is an emulated server - it behaves like a real physical server, and it sits on physical servers underneath the hood - but the difference is that it can easily move between or be spread across multiple physical servers. The advantage of a virtual server is that it does not have to be bought, physically maintained or repaired, it is abstracted from the responsibility and risk associated with the underlying hardware components.
For example, a virtual server may have its files mirrored on several physical servers, so in the event of a hardware failure - it can be rebooted in a different location, most likely with its files intact.
Virtual servers aren’t a panacea, they do still suffer outage, and they’re not immune to increased load either. But they are often protected against other users of the same service consuming more resources, as they’re usually capped to prevent this. Many hosting providers offer a single virtual server to lease, known as a VPS (Virtual Private Server).
Infrastructure as a Service
Some cloud providers have gone a step further than hosting, to offer a range of services that can be termed infrastructure. They provide not only virtual servers, but many other services which can make developing, running and maintaining a website or web app far easier.
Additionally, they have moved toward new cost models which allow for a lot of flexibility. For example, virtual servers are often billed for granularly based on usage, down to the hour and sometimes even the minute or second. The servers can be turned on or off at any given moment, and can be scheduled. If you only need your development servers while your developer is at work, they can be configured to start & stop between 8AM and 6PM, Monday-Friday - resulting in 200 hours usage a month instead of approximately 720.
Some providers of cloud infrastructure include AWS, Google and Microsoft Azure. There are many advantages of such services, and can be particularly powerful if the infrastructure is designed at at the same time as you build your app. These services are well suited to apps and websites which are complex, require a lot of resources, or expect high volumes of traffic.
What is Managed Service Hosting?
As you can see, there are plenty of potential risks with hosting. If you rely on a website or app to drive your business, you need to know that it will always be available to your customers. Managed service hosting is when a provider takes on the task of managing and maintaining the hosting of your web services (your website or app), so you no longer need to worry about managing servers or anything related to the availability of your web service to your users. This is particularly effective when either you don’t have an IT team, or the IT team doesn’t have the capacity to handle web services on top of their on-site responsibilities.
At Pentascape we offer managed service hosting - our engineers are certified specialists in AWS cloud infrastructure and will work to identify the best solution for your needs. Our approach covers:
- High Availability
Ensuring your web services are resistant to failure, often aiming for up to 99.9% availability
- Monitoring & Incident Cover
By actively keeping an eye on your service, when things hit the fan (and they always do) we will immediately start working to resolve the issue
- Pro-active critical infrastructure patches, updates, upgrades and security recommendations
Keeping servers up to date and patched with the latest security releases
- Backups & Failover Ensuring that critical components of your application are backed up periodically and that failovers are in place to ensure high availability for example, your database server should not only be backed up but also have a warm standby, so when it does fail, there is a second server ready and waiting to go with all the up-to-date data as if nothing ever happened
- Disaster recovery plan and quarterly DR review
Even with the best laid plans, disasters happen - DR plans outline what happens when they do and allow your service to recover as quickly as possible
- Technology Roadmap & Cost Review
Are you getting the best use out of your current setup? Can we improve it? Is a new technology available to simplify a process? Have you changed how you run your business and do we need to adjust how the technology works for you?