Real World Experiences of Hybrid IT – Part 4
Welcome back to this series of blogs on my real-world experiences of hybrid IT. If you’re just joining us, you can find previous posts here, here, and here. So far, I’ve covered a brief introduction to the series, public cloud costing and experiences, and how to build better on-premises data centers.
In this post, I’ll cover something a little different: location and regulatory restrictions driving hybrid IT adoption. I am British, and as such, a lot of this is going to come from my view of the world in the U.K. and Europe. Not all of these issues will resonate with a global audience; however, they are good food for thought. With adoption of the public cloud, there are many options available to deploy services within various regions across the world. For many, this won’t be much of a concern. You consume the services where you need to and where they need to be consumed by the end users. This isn’t a bad approach for most global businesses with global customer bases. In my corner of the world, we have a few options for U.K.-based deployments when it comes to hyperscale clouds. However, not all services are available in these regions, and, especially for newer services, they can take some time to roll out into these regions.
Now I don’t want to get political in this post, but we can’t ignore the fact that Brexit has left everyone with questions over what happens next. Will the U.K. leaving the EU have an impact? The short answer is yes. The longer answer really depends on what sector you work in. Anyone that works with financial, energy, or government customers will undoubtedly see some challenges. There are certain industries that comply with regulations and security standards that govern where services can be located. There have always been restrictions for some industries that mean you can’t locate data outside of the U.K. However, there are other areas where being hosted in the larger EU area has been generally accepted. Data sovereignty needs to be considered when deploying solutions to public clouds. When there is finally some idea of what’s happening with the U.K.’s relationship with the EU, and what laws and regulations will be replicated within the U.K., we in the IT industry will have to assess how that affects the services we have deployed.
For now, the U.K. is somewhat unique in this situation. However, the geopolitical landscape is always changing, and treaties often change, safe harbor agreements can come to an end, and trade embargoes or sanctions crop up over time. You need to be in a position where repatriation of services is a possibility should such circumstances come your way. Building a hybrid IT approach to your services and deployments can help with mobility of services—being able to move data between services, be that on-premises or to another cloud location. Stateless services and cloud-native services are generally easier to move around and have fewer moving parts that require significant reworking should you need to move to a new location. Microservices, by their nature, are smaller and easier to replace. Moving between different cloud providers or infrastructure should be a relatively trivial task. Traditional services, monolithic applications, databases, and data are not as simple a proposition. Moving large amounts of data can be costly; egress charges are commonplace and can be significant.
Whatever you are building or have built, I recommend having a good monitoring and IT inventory platform that helps you understand what you have in which locations. I also recommend using technologies that allow for simple and efficient movement of data. As mentioned in my previous post, there are several vendors now working in what has been called a “Data Fabric” space. These vendors offer solutions for moving data between clouds and back to on-premises data centers. Maintaining control of the data is a must if you are ever faced with the proposition of having to evacuate a country or cloud region due to geopolitical change.