Last time, I promised to talk about cloud projects and why some of them are troubled. And by “troubled”, I mean that they’re not always delivered on time, or within budget/scope.
So, here is a five-step guide to improving your chances of success, especially with the first few projects in the cloud environment. To make it practical, I’ve added some tried and tested resources and tips.
1. Take the time to learn. One step at a time
The cloud is a different environment to on-premises. Explore it, learn how to move within it, get to know the terrain, and then run with it.
Until you can get some experience with it, do not take on a big project.
Get your team to work on something they can deliver in one sprint – 2 weeks – and put it into production. Make it both ambitious and engaging, like: “Make this deployment fully automated.” Let them play around with the tools and learn.
It will pay off later. If you want to speed things up, you can bring in some experts, but remember – your team needs the time to learn, to get to know the new terrain before running at full speed.
Pluralsight is an excellent resource to learn, and it is still available for free if you want to learn the basics of Azure. Pluralsight
2. Break the silos!
Separate network and security teams? Get one person from each of them and put them side-by-side with your developers. They need to work together and challenge their assumptions about how things are done.
Isolation might be one of the biggest obstacles to cloud adoption and project success for organizations with established infrastructure and teams.
Break the silos. Allow people to work together and give them something they can accomplish quickly.
3. Understand the cloud model you are going to use
Remember: Cloud computing is not an entirely new technology – it is a new model of delivering computing power and storage, packaged into services. Those services can do wonders for you and speed things up when building your product (be it application or infrastructure).
Each service has its model of:
It is easy to say, “Oh, this looks cool; let’s use it,” only to be bitten by the cost of your product on the next bill.
Speed is essential, but as for learning the tech, take the time to understand the service models you intend to use.
4. Don’t try to reinvent the wheel. It has already been invented
There is a common fallacy when it comes to technology. You can always sense it when someone starts saying: “It is not for us, we are special, and we need to do it differently”.
Sometimes, it is true. In most cases, you will be better off starting with something ready-made. Your problem might be unique to your business, but the technical part has likely already been done.
Here are two quick-start resources:
In the beginning, you will be better off using something that someone has already created. Believe me. Been there, made the mistake of thinking that “we are special.”
5. Iterate and document!
Do not aim for a big hit in just a few months. It is especially crucial at the beginning – ITERATE!
Iterate on the technology, the way you work with it, your team structure, and how you work together in the cloud.
You can make one big mistake in 5 months and learn from it, or a bunch of smaller errors every two weeks, and learn from them as you go.
One common error is to keep these lessons hidden. With each iteration and decision you make in the project, do two things:
How to get started?
Keep things simple!
Simple things are most often omitted as we tend to complicate things. It is important to address the basics right from the start, as it builds a foundation for future engagement.
You need to learn to walk before you can run. This is also true with the cloud.
Already in the cloud and delivered your projects?
What have you learned that you think others can benefit from? Share it with me, and we can compile a list of top best practices for our community.
Read similar articles