There are a whole range of different reasons why you may be needing or wanting to upgrade your server; your organisation may be looking to expand, your old server may have started lagging behind the required levels or you may simply just be looking to build a more integrated solution. The common theme here is – similar to the many reasons for updating your server infrastructure, there are also many things you should be considering to make the process as smooth as possible. Carry on reading to find out 6 of our top-tips on just how you can achieve the most seamless upgrade process possible.
Find out what drivers are supported
One of the most important things to consider when buying your new server is, will it offer you the necessary driver support? It’s important to take a little bit of time to talk to your supplier about whether you will be able to get a hold of the necessary drivers for your chosen operating system, saving you a lot of time and headaches down the line.
You can also do a bit of digging of your own here, where you’ll find many suppliers will offer a whole range of support for different servers. So, by combining your own research with that of a suppliers, you can start to build up a strong understanding of if the server you are looking at is the right one for you.
It’s important to note that most major suppliers of servers in the UK will support Windows drivers, so as long as this is the path you will be taking you shouldn’t experience any issues. However, alternative drivers such as Linux may not be as easy for you to get your hands on. Again, if you get your shovel out and start digging, you’ll find out about the kind of driver support that the server you have got your eyes set on has and will certainly pay off moving forward.
Consider what internal storage you need
Another thing that you should remember when looking at updating your server infrastructure is the internal storage that the server will hold. You’ll find that servers range from the most popular support of two internal hard drives, to servers that have absolutely no internal storage at all.
And it really isn’t as simple as storage or no storage either, the systems out there in the current market include anything from; SATA, NL-SAS, SAS, SSD, NVMe, to 3D Xpoint. Whilst all of this may seem like a range of abbreviations, the differences between systems can really make a difference, so again, a chat with your supplier here goes a long way.
One thing that many schools overlook at first is storage for their CCTV systems. With tighter restrictions on safeguarding and safety within education, it’s vitally important that the necessary services are in place to protect both staff and students. So whilst many servers offer a USB or flash drive storage system, this is simply not enough when dealing with the large amounts of stored data that comes with modern 4K security systems. If you don’t want to lose that data, make sure you have the storage to hold it all!
Depending on what you will be needing the server for, you may not need the internal storage as standard, but it’s best to consider what features you will be needing going forward in the future. What I would say our top-tip here would be is to consider leaving space to expand your storage, after all there’s no point getting something that isn’t going to be suitable for your organisation in a years’ time.
Allow for Future growth
Leading on from my last point comes possibly the most important thing you need to consider when purchasing your new server, and that is allowing for future growth. A lot of organisations tend to assess what their current server needs are, rather than analysing what needs they may have 2 or 3 years down the line. There is little point preparing your server for the needs of today if you are planning on expanding tomorrow. I don’t know about you, but I’d rather be prepared for the next 5 years now, rather than having to renew my server 2 or 3 times in the same space of time!
A couple of points to help you consider your needs for both now and in the future, are:
- How many people access my network now and how many people do I think will be accessing my network in the future?
- Will my new server have the processing power and memory to handle any increase in traffic that I will be experiencing in years to come?
- How much hard drive space do I need for my data now and how fast will this data grow over the next couple of years?
- How many devices are on your system now compared to how many devices may be on your system in the future? (Consider how each user may have a mobile device, laptop, desktop and mobile computer/tablet).
If you take time to analyse these few points as well as any others you have when deciding on your server’s capabilities, you’ll save yourself from getting into a potentially sticky (and expensive!) situation in the future.
Consider CPU support
You may start to see a bit of a trend here towards preparing for the future, but it’s for a good reason. Would you buy a PC today that had specs you need for a current project, but know wouldn’t be able to cope with a project you have got coming up in the future? I wouldn’t either! The same thing applies when you are buying your server, and couldn’t be truer of the server’s CPU architecture.
Take a second to think about how many processor cores you are going to need for your organisations needs right now. Or are your needs processor core based or clock speed dependant? You don’t want to be spending money on a system that doesn’t fit your specific needs, so asking yourself these questions now can be an excellent move.
Plus, you may only need to start out with something along the lines of a quad-core CPU, but this may not be the case in a years’ time. So it’s a wise decision to make use of additional CPU sockets when first purchasing a server so you can upgrade and add to your configuration at your convenience, rather than updating the whole thing again.
Utilise Redundant components
You don’t need me to tell you how integral a server is to your organisation, so what happens if something goes wrong? There are a number of reasons that a server may experience downtime or malfunctions, so instead of sitting there waiting to see if it happens to you, why not install redundant server hardware to act as an effective fail-safe?
Redundant hardware is what is installed as a backup of sorts, an extra set of your vital components that runs idle and will kick in, in the event of any sort of failure with your core components.
It’s best to think of every eventuality if you are dealing with important data, for example; if your server has two power supplies, when one fails there is still one running so things don’t go offline. Another top-tip would be to ensure you provision backup storage, which will be used in the event of a catastrophic failure.
It’s the supposedly ‘little things’ you can do now which will ensure that your server runs as smoothly going forward.
Consider a cloud server
I’m sure by now that you have heard of these trendy new servers and they can be temping, I mean after all, you get to delegate all the upkeep and day to day running of the server to someone far, far away.
But is it a worthwhile investment taking all of your vital information, programmes and services and sending it to live with somebody else? Well there are a whole host of advantages in cloud computing, namely:
- It’s cost efficient – due to the fact you are essentially investing in a service not a server, you won’t be forking out cash for hardware and upkeep, it’s all ran and maintained by the third party.
- Unlimited storage – you also don’t need to worry about running out of storage capacity either, a cloud provider may charge you for any extra data stored but you can customise this to the needs of your organisation, meaning you aren’t paying for something you aren’t using.
- A hassle-free deployment – perhaps most importantly, a cloud server service can be up and running for you within minutes. This time can vary depending on the needs of your organisation, but either way, it’s a lot faster than whole server installs.
Okay, so cloud servers sound very tempting, but you’re probably thinking there must be some disadvantages or everyone would have one? Well, there is unfortunately and it all comes down the third party.
With a cloud server, you are essentially putting your sensitive information in the hands of a third party who are then responsible for it. A bit of research and digging can help you out significantly here however. It’s 100% worth checking if the provider you are looking at is both reliable and not prone to repetitive technical issues. I know everyone has technical issues from time to time, but if a provider appears to be having them repeatedly, you are just going to be seeing a lot of downtime, therefore limiting productivity.
So, that was our 6 top-tips for your next server upgrade, but don’t stop searching there! Whether it’s a dedicated in-house server or a cloud based solution that you opt for, the biggest thing for you to remember is – prepare for the future not the present. You may find a great price on a server that copes with your needs now but can’t cope with the demands going forwards, or a golden offer on a cloud service that gives you what you need now but will charge you excessively when you need to expand.
To help you decide, why don’t you check out some more details about the advantages of going to the cloud or some of the latest developments coming from Intel® in the form of the Purley Platform. If you need some more advice discussing your next steps, or any of the latest information on anything server related, our dedicated infrastructure team here at Stone will also be happy to help!