Before I get carried away with my favorite bard, let me make something clear: this is not a what-to-do or a when-I-do for moving to the cloud. This is a set of pointers on how-do-I-move to the cloud and more importantly, covers specific, real world use cases for moving to the cloud.
Coming to the bard, when Hamlet bemoans “to be or not to be”, what he really is saying is that the uncertainty of his predicament – is killing him! While he thinks an easier way would be to end everything (including his life), the alternative seems just as bad!
So, why am I bring this up? Simply because I’ve heard the exact same refrain (in a cloud based context, of course), from CIOs, IT Heads, VEEPs of Infrastructure, CFOs/CEOs across enterprises, SMEs, business domains and even my local kirana shop (corner shop for you non-indie speakers). Ok, so the last one was thrown in just to check if you were following my train of thought…
Like Hamlet, businesses grapple with maintaining status quo (the easy way out) and losing the race by not leveraging technology; or, explore and move to newer technologies that might give them competitive edge, but entail exponentially great risk in the short term (the more difficult choice).
We have all heard the pro-cloud arguments – it’s cheaper, it’s elastic, it’s plastic, it’s the best thing since sliced bread, it’s the second coming (among other things) … basically, your standard-issue, new tech-babble!
But is it really (and I mean seriously) a game changer?
Having been part of multiple arguments and debates – both, from the customers/ consumer standpoint, and as a multi-platform cloud service vendor (yes, I occasionally moonlight for the dark side) – I wanted to put a few things in perspective based on experience batting for both sides; that might, just might help all you business (and technology) heads out there in simplifying your decision to move to the cloud.
So, read on for some real-world, business use cases for considering the cloud.
When you have stable workloads, fixed users, no spikes and security is a major concern (read enterprise grade, commercial-off-the-shelf apps (SAP, Oracle, CRM, etc…))
– don’t move to the cloud. Just don’t. Unless it’s time for a refresh and most of your users access said applications via the internet (which usually will not be the case).
Then, and only then, consider a cloud based setup with Compute, Storage, Network, Security (preferably with CDN) for this use case and only if you have a good managed services partner or an in-house, cloud certified expert to guide you through the process.
Or, move to a cloud setup offered by the application vendor themselves (which most of them do today anyway).
Key Takeaway: You definitely can save money by moving to the cloud, but only if your application infrastructure is essentially internet already or if the cloud setup is provided by the application vendor themselves!
When you have wildly fluctuating workloads, varying user base, major spikes, and security is a major concern
– such a usecase just begs to be moved to the cloud. Consider it a given that you should look at least moving the worst of your spiking workloads to the cloud; maybe with a virtual private cloud (VPC) setup (given your security concerns).
Key Takeaway: You will save some money, by moving to the cloud. But only after a careful assessment on extending security to the cloud and this might involve re-jigging parts of your application! Again, a managed services provideror an experienced multi-platform, cloud partner – preferably with a security team– would be ideal!
When you have steadily increasing or wildly fluctuating workloads, unlimited users (see, I get better as I write), peppered with spikes and security (still) a major concern
– move to the cloud. Right now. I mean seriously, there should not even be a doubt. Elastic workloads are a natural fit for the cloud.
Also, ensure that you sign up with a good managed service provider (just to reduce headaches you know. You don’t want “ghosts-of-your-deployments-past” to visit you) to guide you from the architecture to the migration and on-going management of your setup. So, ensure that you look at partners that have experience managing such environments in your specific tech and business domain.
Key takeaways: You will save a lot of money and will also scale based on your requirements – and that too, seamlessly (if you have the right architecture)!
When you generate large amounts of data, with unlimited or limited users, featuring spikes and troughs, with security needs that are built-in (especially compliance)
– go on the cloud. Why you say? Cause you’ll love the sound of loose change (cost savings) in your bank account every month. Also, there might be some surprises in a couple of months (the months that you roll out optimized code that reduces load on your production system or leverages the underlying cloud platform’s storage options properly).
Christmas just might come early for some of you.
Key Takeaways: Lots of cost savings! Especially, if you can get your application vendor to optimize code for cloud deployment (based on specific features that your cloud platform provides), along with a good security vendor (or maybe, get one who does all). Big data is made for the cloud and there is are tons of studies to explain the use case out there.
And finally, looking at deploying an application for an industry that needs compliance – HIPPA, PCI, etc… etc…
– Just deploy the said application on the cloud. And voila! You get instant compliance at the infrastructure level with practically no extra cost (other than ensuring your application is compliant from a software perspective – which you should have done anyways, if you are creating software that is specific to an Industry).
Key Takeaways: Long term cost saving + compliance + future proof solutions! Need I say more?
In closing, all I would like to say is that while the big guys (read cloud platform providers) keep telling you how their cloud is the next best thing after hot air (pun definitely intended), get your head out of the clouds and take a serious look at your workload and it’s characteristics. I promise you will find it much simpler to make the decision to move to the cloud after that simple exercise.
An added benefit – you will not roam the halls muttering under your breath – “to cloud or not to cloud, that is the question!”… 🙂
Agree / disagree or dismiss this as the ranting’s of a Shakespeare-loving-techno-business-geek; but please leave your comments/thoughts. Especially, if you can add some more real-world business use cases!