The Cloud, not just someone else’s computer
If you have spent time on Twitter, Facebook, or LinkedIn recently then you have most likely come across one of the “There is no cloud, it’s just someone else’s computer” images that are trending right now. The statement isn’t new, but this hasn’t stopped the images from spreading like wildfire and stirring up discussions on social media lately.
While people seem to enjoy poking fun at their technology illiterate friends who apparently believe their family photo collections are stored in some magic cloud out there, it is a statement worth reflecting on even for the ones that eat ones and zeros for breakfast.
If the cloud is just someone else’s computer, why are businesses falling over themselves to push more and more into it? Surely business divisions have always seen their IT estate as someone else’s, be it their internal IT department, or belonging to one of their outsourcing partners. If so, why isn’t IT outsourcing generating the same buzz?
The answer is because the cloud isn’t just someone else’s computer, it’s a lot more than that. When AWS burst on the market in 2006 they offered fairly basic compute and storage services and early adopters focused almost exclusively on a “lift & shift” strategy where they migrated off an ageing on-premise infrastructure to a similar setup in the cloud. Even if this can provide organisations with more flexibility at a lower cost, it is more an evolution rather than a cloud revolution.
However, over the last decade Amazon and other market leaders have worked relentlessly to add tools and services sitting around their core offerings that go far beyond what any internal IT organisation is able to provide. While you can argue that these tools and services also just run on someone else’s computer there is more to the story in the same way you wouldn’t consider it just a car change if you managed to trade in your old Ford Fiesta for a brand new Mercedes leased on a cheap hourly basis with all the latest mods plus 24/7 breakdown coverage thrown in for free.
What tooling do you need?
It is relatively straight forward to move off an on-premises infrastructure to a similar setup in the cloud, but it simply puts your mess in the cloud. If you’re looking for more fundamental changes then you need to go beyond this and adopt a Cloud-First strategy. A Cloud-First strategy means architects, procurement and business leaders all work together towards designing, building and operating new solutions in the cloud.
This is a potentially huge change to how organisations currently think and will have a significant impact on processes across business, development and operational areas. So where do you start and what tools and services are right for you? Organisations have different needs so there isn’t a simple answer, but the information below may set you off in the right direction.
DevOps
Organisations that undertake development are likely to end up having conversations around DevOps – a phrase used for a type of agile relationship between Development and IT Operations. While the principles behind DevOps has nothing to do with cloud specifically, no one can deny that its recent rise to fame is in large parts down to cloud vendors providing programmatic access to their services. With the ability to fully automate and programmatically configure large-scale production environments in minutes the lines between Development and Operations has “blurred” considerably.
AWS, Azure and other cloud vendors offer a wide range of tools and services that can help deliver on the promises of DevOps, but it can be challenging for an organisation to work out what is best for them. If developers predominantly use Microsoft technologies such as VisualStudio and .NET then obviously Azure’s service offerings stand out, but often it isn’t as straightforward.
To avoid vendor lock-in it may be wise to look beyond the vendor’s own offerings. Configuration management tools like Ansible, Chef and Puppet come with “plug-ins” for all major cloud vendors, which makes it easier to move between them, or to handle multi-vendor setups. Many tools are also open source, subscription based and well-supported, meaning there is really no excuse for not exploring them.
Big Data
Today, organisations are not only experimenting with big data and analytics, many have already started to see real results and are expanding their efforts to cover more data and new models. Like DevOps, big data has nothing to do with cloud specifically, but it is the low cost and high flexibility of cloud solutions that have made it a reality for many organisations. With the introduction of SaaS (Software as a Service) solutions such as AWS Redshift and Google BigQuery the barrier to entry has become even lower with organisation seeing huge improvements over traditional data architectures, a price comparison alone can suggest order-of-magnitude improvements.
Security
With cloud, security is often a hotly debated topic and there is a lot of scepticism about storing sensitive data in the cloud. While there are regulatory challenges around where data is physically stored, particularly if data centres reside in a different country or continent, the general perception that cloud is less secure than on-premise solutions is misguided. Yes, there have been some high profile data breaches, like Apple’s iCloud breach, but they are rare. The likes of Amazon, Google and Microsoft have some of the sharpest minds and biggest budgets at their disposal so it would be foolish to think that your in-house IT department provides a more secure option.
Some organisations, particularly within financial services and governments, do have very specific needs where SLAs and security controls from the more popular cloud providers aren’t stringent enough. In these circumstances it may be necessary to explore cloud vendors such as Skyscape that can provide services that go beyond the already exceptional security levels provided by the big players.
Final words
Adopting a Cloud-First strategy is a big step for most organisations and will require some careful considerations, but that shouldn’t hold anyone back from exploring these tools & services now. Too many organisation apply a “No we don’t allow anything in the cloud” policy and that needs to change.
Let development and operational teams experiment with AWS, Azure or one of the other cloud platforms. The associated costs are very small and as long as no sensitive information hits the cloud the associated risks are negligible. Allow developers to “hook in” their favourite tools, start explore configuration management solutions like Ansible, Puppet and Chef to see what works and what doesn’t. There is no need to be too prescriptive at this stage and any findings, good or bad, will help shape the foundation of your cloud strategy. Let the revolution begin!