Part One – What is cloud anyway? By Russell Warne CCO Kaskade
Whatever industry you’re in, you tend to forget that what seems obvious to you is Greek to someone who isn’t in the same game. In IT we’re probably guiltier than most, and even more so when it comes to ‘cloud computing’ as its definition differs depending on who you talk to.
Before we break this down further, this guide focuses mostly on Infrastructure as a Service (IaaS) – essentially the servers, storage and networking part of things – and touches on other services like PaaS and SaaS. If you’re not familiar with those terms you really don’t have to be for any of this to make sense, but you can find out more in this article or listen to our podcast here.
In addition, you should be looking at the benefits, not worrying too much about the definition, but it’s still important to know what your provider is able to deliver and not get muddled with technical explanations. To simplify things, we want you to think about cloud as just lots of servers and other hardware in a building, or datacentre. Many would say that’s like calling town or city a ‘bunch of buildings and roads’ and doesn’t allow for the true distinction of many cloud services, but as far as infrastructure goes it’s actually a good analogy.
We have Villages, Towns, Cities and Mega-cities. Each has it’s own benefits and not everyone wants to live in the Mega-cities but it definitely has its advantages. So let’s dive into the analogy.
Servers hosted with a provider – often the datacentre within your reseller or space your reseller has taken with another provider. Like a village, these would have limited options and services, smaller hosters can also only provide a limited number of compute, storage and networking resources. Like a village only has one medical centre, smaller hosters will only be in one location and so your workload will only run in one place. Roads (networking) will have limitations and whilst there’s police station, your security is limited to the needs of that small village/operation due to the running cost of extensive security options. There’s a restaurant and some shops, but not a lot of choice and most likely no sushi 😉. Similarly, you’ll only be able to run a few types of operating system and you’re not going to be able to do much more than run those servers as they are now. If your staffing needs grow for a project, expect your office to be crowded for a while, like your compute/storage other requirements may be crammed closer or over capacity in a small hosting environment. It’s a great place to call home, but change is slow, options are limited, and if there’s a fire at the grocery store you’re going to be hungry for a while. Most businesses have such a high dependence on IT that this kind of option presents quite a lot of risk.
Here you are dealing with companies that have focussed on providing enough servers and storage to make that their business focus – often using partners like resellers to on-sell the infrastructure to you, the customer, rather than go and deal with the end-consumer directly. They may even have hundreds of servers and other hardware, probably in more than one location, and like a town there’s benefits to that – like maybe an additional hospital (you can run your workload in two places in case something goes wrong in one of them), more choices of shops (greater breadth of server and operating system choices) and with more people, the roads have more capacity and better control so things flow better under times of increased demand. Of course, if you have an accident there are limited options and you could have congestion (you’ll be getting the analogy of with network response by now!). If you need extra office space, there’s some extra capacity and, just like you could get bigger or more servers, you can move offices or maybe hire temporary space for times of increased need only. However, you’re likely to have to sign a lease for that so you must do so carefully. Same with servers – you’re likely to be entering contracts for months to years. Security becomes more important and there are multiple means to secure things. In general, things are slicker and centred around having more people live there easily.
It’s sometimes difficult to make a distinction between a town and a city, but typically a city (and metropolis) could be broken down into multiple towns and suburbs, and the infrastructure that supports the increase in people starts to take on specific characteristics. For example, you may have a subway, and access to facilities that can move large numbers of people in and out (like big airports) become mandatory. This is where we put the large, national and some international cloud/hosting providers, like local telco’s and big internet service providers. They have multiple locations (suburbs/towns), and a significant number of options in terms of platforms to cater for a wider variety of needs. Those airports are big investments, but trade is big enough to justify it, just like significant investments in connectivity, security, fire prevention and guaranteed power are now feasible and used to attract bigger businesses with more stringent IT requirements. Cities can accommodate growth of company’s requirement’s quite easily, but there are the usual commitments (like term leases), albeit there’s far more flexibility. And like you’ll find a myriad of services in big cities you wouldn’t find elsewhere, big hosting and many cloud providers offer a breadth of services not found in smaller operations as they have bigger markets, access to skills, procurement power etc.
There’s a classification known as a megalopolis – a grouping of multiple large settlement units each of which might have 10 million people. The jump from City or even Metropolis is a big one and we’ve used this gap on purpose – the jump to the Hyperscale Cloud Service providers is so large that it’s unlikely any contenders will be able to reach this scale. For example, a single AWS datacentre will have more than 120,000 physical servers, supporting millions of workloads. There are 55 of these around the globe, with plans for 12 more already published. They design their own networking technology, run their own power substations and have the ability to customise their infrastructure to their customer’s needs. And while many of our human settlement cities have grown through ‘sprawl’, the hyper-scale cloud providers have started with plans to serve millions of customers, so everything they do is to provide services that people want at massive scale. In a way, they are the theoretical ‘perfect’ city, drawing on the aspects of all types of infrastructure to provide something close to an ideal. That may sound like we’re pitching a Panacea, but the hyperscale providers really do have an edge in that the ‘basics’ of things like compute, storage and networking are the things they do in their sleep, while their focus is creating the kind of things people will move cities for. Always on infrastructure, global reach, self-service, huge bandwidth, advanced technologies accessible by all citizens – these are some of the benefits small and medium sized businesses can access.
So, wouldn’t a small business be better in a village or town? Well, to really stretch the analogy, small businesses can take advantage of all the things mega-cities have to offer, but still live in the country. You get to access the markets and technologies that the world’s biggest companies do, at the same price, and can take as little or as much of it as you need. Once you’re in the cloud, your infrastructure become malleable and accessible from anywhere. In our view, there is no reason not to benefit from the advantages that these Mega-cities have to offer. It’s not only price, but service and product selection as well. We’ll dive into what this means in real terms in Part two.
If you would like to accelerate your business success in the cloud that over 1 mil customers are using, then take advantage of Kaskade’s special no risk business offer. Click here for more info.