Shifting business applications and data into the cloud delivers a step change in multiple ways. There are the financial benefits of removing CAPEX from the budget and putting it into OPEX instead. There are the disaster recovery features and business continuity advantages. And, not insignificantly, there is the advantage of always having the latest advances in technology at your disposal. Without the onerous upgrade overhead.
Look at Amazon Web Services or Microsoft Azure as cases in point. These giants of the cloud industry provide hundreds, if not thousands, of services which can be accessed with the ease of a mouse click. Those services – and capacity – are available on-demand. Need a database? Click. Need computing power? Click. Need machine learning? Click click.
Compare and contrast with on-premise procurement. Need a database? You’ll need some hardware to run it on first. Put in an application for CAPEX to the board, then wait. Got the OK? Good, now set up the tin in the datacentre; you’ll need some engineers to get it right. Oh, and operating system licenses. And then you’ll need to buy some licenses for the database itself, and those are so expensive that they had better have appeared on the initial CAPEX request.
Once you’re up and running, which can take weeks, months or even years, you’re effectively stuck with the technology you ordered at the very start. Want to upgrade it, because you deployed SQL Server 2008, but there are features in SQL Server 2016 that could seriously benefit the business? Not so fast. Start the whole process again. Oh, and your ERP system isn’t yet compatible with 2016? Maybe just forget about it.
And consider this. Let’s say you got the green light on your project, but it didn’t deliver the goods in practice. Egg, meet face. Embarrassing, to be sure, but also expensive. And what becomes of that hardware and those licenses? Shelfware is real. You don’t want to contribute to it.
With cloud software, there’s no need to countenance that onerous and quite frankly outdated nonsense. Keen to try something new? After all, machine learning is all the rage these days, what could it do with your data? Just access the service and give it a go. Want to spin up a new database to test a few hypotheses? It’ll cost a couple of hundred bucks to get all the server power and storage capacity needed. It works? Great, expand away, on demand. Value creation more accurately matches cost.
Constant upgrades (and no more Patch Tuesday)
The pace of technical development in the cloud environment is exemplified by Amazon Web Services. Let’s take one aspect of its EC2 compute service as a case in point: Microsoft SQL Server Standard Edition. Not only has the price of this service been reduced 62 times, but new features are constantly introduced.
This isn’t peculiar to EC2. It isn’t even peculiar to Amazon Web Services – the same can be said for Microsoft Azure or Google Cloud.
Compare that with the on-premise reality. Software 5 years old isn’t unusual. You’ll probably have some 10 or 15 years old; Windows XP, for example, is the third most popular operating system in the world. Ahead of Mac OS-X and all flavours of Linux; needless to say, you’re missing out on a lot of software capability if you’re stuck in the past.
And, unlike on-premise software, every time an innovation takes place on the platform, every single customer can access it immediately. No need to perform a physical upgrade. No need to wait. No need to change anything or fiddle with hardware. Because it is all in the cloud, the latest advancements are constantly and immediately within your reach – and most probably, given that history of price reductions, at a lower cost.
That’s turned how the software industry works almost entirely on its head. It means ‘technical’ innovation in software, which could once deliver a competitive advantage, is now freely available to all. This allows you to focus on differentiating elsewhere – and that means better products and services for everyone.
Even the really fancy stuff is available as a service. How about automatic SQL Server database tuning on Microsoft Azure? Pretty advanced – get it at the click of a button. Machine learning? It’s yours, for 42 US cents per hour. No need to ask the board for the OK on that one.
Oh, and Patch Tuesday? That one day a week where you’d have to make sure all systems were secured and up to date with the most recent fixes? Forget about it. When its server-side in the cloud, that happens automatically and incrementally. There’s no lag between the release of the latest patch and when you install it.
Computing and data storage have the marvellous property of becoming cheaper for more as time goes by. When you’re in the cloud, you don’t only get the latest goodies available from processor, software and storage hardware vendors, you get it for less money. In the on-premise world, what you buy is what you get, end of story.
Illustrating how quickly price and performance change, we specified the very latest processor model for a client, provided Microsoft Azure. Three months later, a newer generation was out, so we switched the client over. They got more RAM which made their application perform better, and that was for two hundred bucks less. It’s economics like you’ve never seen before – and the upgrade was carried out, you guessed it, with a mouse click.
The bottom line
There is a bottom line here and it is a pretty simple one. Cloud computing, particularly for enterprise applications (where practical) delivers advantages which are hard to ignore. The latest technology at your fingertips, always, is a big deal. Upgrading to it with zero impact is a big deal. Getting more for less is counter-intuitive, but that’s how it works online.
So, the bottom line is this. It’s time to see where and how cloud can work for you.