In an earlier post I made mention of the fact that the Microsoft Cloud OS vision is based on the principle of meeting business needs rather than managing technology limitations. This vision is based on a common maxim on the minds of all IT Pros as they evaluate and implement technology to enable their business: Make technical decisions based on the priorities of the business.
This mindset is why IT Pros worry about vendor lock-in. At its most basic level, this fear is about putting the future of their company (and their own personal future) in someone else’s hands. I admit that it is obviously a little strange to hear a cloud service provider, like Microsoft, tell you how to avoid lock-in – but this kind of freedom and opportunity to choose is one of the single most important elements of cloud architecture and our Cloud OS vision.
Locking Down Lock-in
Before I get into the technical aspects, I want to emphasize some additional reasons why it’s so critical to avoid lock-in, and why I believe lock-in is not the same thing as having a single vendor.
My definition of vendor lock-in is a state where your technology options become limited (or your service needs are not met) and overcoming these limitations is either cost prohibitive or technologically impossible.
With this definition in mind, choosing a service partner is an exercise in risk assessment, in particular:
- Will my cloud vendor work to ensure that my pricing expectations are kept stable during the lifetime of my service?
- Will my cloud vendor be able to meet the service levels that I require, e.g. system performance, availability, and service desk?
- Will my cloud vendor make continuous technological innovation a priority?
Maintaining the power of choice is the constant throughout all three of these questions. That choice extends to both the provider and consumer of the service.
A lot of my thoughts on pricing are already addressed in my post about the Virtuous Cycle of cloud computing, so in this post I’ll focus instead on the other two primary risks, and I’ll examine how a Hybrid environment allows you to focus on delivering value and innovation for your organization rather than constantly working on service migration strategies.
For the purposes of this post, I’ll look at what I believe are three most important areas to avoid lock-in, and what you can do to stay proactive about your cloud choices and cloud flexibility:
- Hardware & Fabric
Any cloud provider recognizes (even if they don’t want to admit it) that our customers are going to move the location of their services as they see fit. Microsoft has architected its Hybrid Cloud approach in such a way that makes us the only vendor with a three-cloud solution specifically designed to ensure that you can deliver solutions that are able to take advantage of all three cloud settings, i.e. on-prem, with a service provider, or publicly in Windows Azure.
Each of those clouds has specific characteristics that are desirable across a wide set of areas (e.g. price, control, features, etc.), and choosing a vendor that only provides one or two of these cloud environments won’t allow you to optimize across those areas – and this is what increases the risk of lock-in. This is why the multi-cloud, Hybrid approach to IT is so important. In a sense, a Hybrid Cloud prevents you from being “locked-out” of the specific benefits of each environment. On any given day, you may not necessarily require every aspect of each cloud, but I believe there’s a huge value to having these functions available when the need arises.
Another aspect to consider is how much can change during the lifecycle of a deployed service (e.g. modified regulations, changing laws, natural disasters, fluctuating markets, new business priorities, etc.), and our vision is to enable transitions between clouds without having to worry about limited support due to proprietary technologies. These transitions should be able to occur without the pain of having to completely re-architect your solutions. I am proud of the fact we’ve accomplished this in a unique and frictionless way by supporting open standards, creating consistency between clouds, and providing you with the ability to transition your services to any provider in the Microsoft Cloud OS. In this scenario the burden is squarely on each and every provider to constantly innovate in order to keep you as a customer. You get your service where you want it – period.
Hardware and Fabric
Microsoft has learned a lot from the development, delivery, and operation of Windows Azure, and we believe in sharing our cloud innovations to foster more advanced datacenters and the adoption of cloud computing. A couple recent examples include our joining the Open Compute Project, as well as our announcement that Dell is supporting Storage Spaces. We are dedicated to transforming datacenters through the power of software.
Storage Spaces is a great example of this: A way to deliver high performance, high quality services on-prem (or as a service provider) while keeping expansion costs under control.
It’s also not all about Microsoft products. In recent years we have worked hard to ensure that the Cloud OS is best place for Linux. To do this, we’ve made the necessary investments to ensure that our Linux integration keeps up with our innovations for Windows. To ensure that Linux is a viable choice on Hyper-V, we’ve added Dynamic Memory and video driver support.
Another great example of our commitment to enabling choice can be seen in our approach to Network Virtualization – specifically, Cisco’s integration with Hyper-V Extensible Switch. Our work with Cisco has enabled an open, extensible, and standards-based approach that has created great partner ecosystem momentum for our networking solutions. Elsewhere, many of the top networking and merchant silicon partners are building solutions on top of Windows Server and System Center. Our vision, in every aspect of these cooperative efforts, is to innovate as much as we can in the places where we are responsible. These ongoing efforts never lock-out the innovations happening elsewhere the industry.
Many of the most exciting innovations in the Hybrid Cloud focus on what we (as a provider) and you (also as a provider), can offer the frontline consumers of our services, aka Developers.
There is no shortage of available information about how Microsoft has changed as a company and integrated openness and choice into the cloud, and we recognize that there are specific Operating Systems, Databases, Applications, and Languages that need to be part of any comprehensive service catalog. Our goal is to create a space for developers to invent and innovate without obstruction and with the tools they know how to use – but we want to do it in a way that can be managed, operated and supported throughout the entire stack.
Earlier I noted some of the great work we’re doing to make sure Windows and Linux can work together in the Cloud OS. Looking further up the stack, we are also working to establish a gallery of pre-built VM images that are optimized for the Cloud OS and validated by their respective vendors, such as Oracle.
We create App templates for use in Azure, and make them available on-prem or through a service provider with Windows Azure Pack. This way, if you want to offer Joomla or Drupal to your developers, you can add them to your service gallery. Another important contribution towards enabling stability and choice is the work being done at the VMDepot – a community-driven gallery of VMs ready for the Cloud OS and, in many cases, supported by 3rd party companies.
With the Cloud OS vision, we are dedicated to enabling choice for both providers and consumers of services. When developing a hybrid cloud strategy, evaluate your options with lock-in avoidance as a top priority. Ask yourself whether your strategy puts the services you provide or consume in a position that forces you to compromise your infrastructure or your organization.
Ultimately, avoiding Lock-in is about choosing a great partner who is constantly innovating, offering you a great price, and empowering your organization to become better.