Andrew May, Cobweb Cloud Solutions Architect, answers the question What is the cloud? – and where can I see it! …
We often hear phrases such as “migrate it to the cloud”, “saved in the cloud” or “use the cloud to store it”. But what does that actually mean? Where or what is ‘the cloud’?
To many people, the cloud is some sort of abstract construct that exists somewhere between the various planes of existence. I’ve seen various jokes in where the punchline is “nobody understands the cloud, it’s a mystery!”. For those new to the IT industry, the concept can seem a little difficult to grasp.
One of my favourite cloud related memes is “What if I told you there is no cloud, it’s just someone else’s computer”. And that is exactly it, the cloud is a collective term for any IT services that run on someone else’s computing hardware and usually they’re accessible via the internet. Think back to the early days of dial-up internet – your Internet Service Provider (ISP) normally offered a free email address, a few megabytes of website storage, etc. bundled in with your internet connection. Those services were hosted on computers the ISP (or some third-party) owned and these days would be referred to as cloud services.
So now we know that anything “in the cloud” is just running or stored on someone else’s hardware, where is that hardware and is there anything you should consider when using the cloud or choosing a cloud provider?
IT hardware is typically stored and operated within dedicated, specialised buildings known as datacentres. These buildings are filled with metal racks into which the hardware is installed. The racks are cabled with redundant, high-capacity electricity supplies that are backed up by batteries and generators to ensure continuous operation even if the national grid has an issue. There is air conditioning to keep the hardware at the right temperature and humidity. There are communications cables to wire everything together and then out to the internet via various different providers to ensure availability, even if one connection fails.
These datacentres are real places and if you’re determined and can find out where they are, you can go and see them, so they must be secure. They’re often non-descript, grey buildings with few windows, surrounded by high fences, lots of CCTV and sometimes intrusion detection such as Infrared detectors. There’s also a strong security staff presence, usually with patrols. All this keeps the building secure from unauthorised people getting in.
People do have to work in these buildings though and protection is ensured through vigorous background checks for everyone who needs access to ensure they don’t pose a threat and once verified on entry, the use of key cards to ensure those who are allowed in can only access the areas permitted.
So now we know the cloud is made up of datacentre buildings, how do we know where they are and does that even matter? We don’t need to know the precise locations and indeed Microsoft does not publish this information for security reasons. But we need to know roughly so that we can ensure our data is being stored and processed in line with our requirements. Data sovereignty is important for many organisations who must ensure privacy is maintained, local laws are abided by, and regulatory compliance is met. To build robust, resilient IT services in the cloud we also need to know that redundant components are geographically separated.
To address this, Microsoft groups their datacentres into regions, which allows them to define an approximate location without giving away anything too specific. For example, the “UK South” Azure region is described as being in London, UK. London’s is a non-specific location given its size but is detailed enough to let us know that it’s in the geo-political boundary of the UK and subject to UK laws.
So, when you put something in the cloud, where does it go? Well, if the cloud is Azure and you choose UK South, the answer is that it goes in a datacentre building, somewhere in London, on hardware Microsoft owns.