At re:Invent 2019, AWS made no less than 75 product and feature announcements across the full six days – that’s more than ten announcements per day. Google went even further and made a hundred announcements during their three-day Google I/O conference in Spring 2019. The speed of innovation is staggering, and this rate is here to stay. For those who have committed to either of these platforms, this is good news even though they might not benefit from all the innovation, only to some related to the services they are consuming. The problem with this is, however, vendor lock-in. What does that mean for you?
At first sight, the rate of innovation is good news. Continually improved services, lower prices, better performance; what's there not to like, you may rightfully ask? After all, you get these benefits for free - or do you?
Looking into it a bit deeper, a pattern of innovation emerges across all hyper-converging cloud providers - that's Amazon, Microsoft and Google. Let's quickly summarise what these are:
- To varying degrees, the 'Big Three' are relentlessly customer-driven: Innovation happens where there is high demand from customers
- They are also profit-driven, hence innovations are prioritised not only by customer demand, but also by profit potential
- A huge customer base has the potential to drown out individual or smaller group requests
- Providers are naturally driven by customer retention; therefore a "golden cage" strategy promises good results and retention rates
A golden cage ... is still a cage.
To be clear, we are not unequivocally advocating against exploiting all the options available for you to make your cloud-based infrastructure deliver the best business value to you.
What we are positively advocating for is for you to make a conscious and informed decision for one or the other option (or anything in between).
A common misconception about modern cloud providers is that they are merely providing compute, storage and networking resources as a service. While this is still true, and form a very significant part of their business still, providers heavily push for what is called Anything-as-a-Service (or XaaS for short) that goes far beyond PaaS (Platform-as-a-Service) and SaaS (Software-as-a-Service).
For instance, if you decided to contract AWS for a database service, you have a number of choices to select from:
- Fire up a VM (or a container) and deploy, configure, and manage any type of database yourself
- Choose one from the 10(!) different AWS-managed databases
The tricky, and much less obvious issue at hand here is less of a data portability and application migration; this is still an issue and an often overlooked cost. The biggest issue here is that of interfacing and integrating with your operational management tools and interfaces. While there are many similarities across cloud providers, there are too many differences in approaches and interface usage which makes switching between providers an often daunting task. Hence, the cost of migration is the true golden cage you buy into.
Are you happy with the realisation? Are you happy in your cage?
Cost of migration vs. cost of independence
Technically, migrating from one cloud provider to another is a solvable problem. The prohibitive factor of migration is the cost attached to it. Therefore, to evaluate the best course of action and eventually the architecture of your IT estate, you will have to assess how important several fundamental elements and features of your IT infrastructure will have to be.
For example: Any reputable data centre provider maintains redundant and independent providers for power (which often includes UPS solutions such as diesel generators), cooling, connectivity, and other core facilities and utilities required to operate a data centre at or very close to true 24/7 availability. How would you evaluate a data centre provider that states something like, "We never had an issue with connectivity, we don't need a second provider for that."? But why are organisations still happy to trust and rely on one single cloud service provider? Why do organisations apply proper risk management and diversification to all other areas of their business but happily put all their eggs into one basket when it comes to the cloud or IT in general?
Unless there are very good reasons and proper mitigation plans in place, any good continuity, contingency and disaster planning must include the possibility of switching from one cloud provider to another. And the more features and unique approaches you use from your current provider, the harder it will be to switch should you ever need to.
As stated before, it is absolutely fine if you are happy with staying with your current provider, given that you made an informed and conscious decision. If, however, you prefer or need choice and independence, you should probably look to re-evaluate your strategy and change your IT architecture accordingly.
Digital Craftsmen remains a consciously independent managed cloud provider, and we are happy to help re-evaluating your IT strategy, migration plans, and digital transformation. Book a free workshop with our ISO 27001 and Cyber Essential Plus accredited expert solution architects today.