Join David Linthicum for an in-depth discussion in this video DevOps example, part of Cloud Architecture: Advanced Concepts.
- [Josh] All right, let's look at an example of DevOps as related to cloud architecture. So, keep in mind your DevOps processes may vary a great deal and this is also based on what we talked about in the tool video that we'd just previously did. So, understand that you need to take this as a baseline of understanding and that you need to create your own process or DevOps process around these sorts of things. So, again we have continues development, continues integration, continues testing, continues deployment and continues operations, continues improvement.
So, what this look like is typically those who's going to be people and let's take a smaller organization like a product development organizations. So, we may have six or seven developers that are doing the continues development. So, they obviously take the requirements that are coming in from the customers and they have a direct link to the costumers. They can actually get it to developers and talk to them about what their needs are. Is it _ unique concept and they're able to make changes and they're able to do some testing, even some unit regression testing on their own.
And once they're ready, they hit the button and release the software. And goes through a build and goes to continues integration. So, we make sure it's syntegrated with the development resources with the database resources, everything we just talked about and once it passes muster there it goes to continues testing, we may have all sorts of testing and that could be a one person who's in charge of doing that because it's typically going to be and they're going to be in charge of the tool and not necessarily in charge of actually doing the test and so they're in charge of writing the tool scripts and the testing scripts to do the testing for the system.
And then from there once it passes the testing and it may not it may actually go back to the developers for some redo, if the testing doesn't work and the developers will get this back instantaneously by the way which is very cool, so in other words they hit a button, it goes off to continuous integration, that may take 10, 15 minutes, and it goes to continuous testing if there's a problem the developers will get an email back as to what happened, what failed, what needs to be changed, and they can make changes to the system and then resubmit it to continuous integration so that loop will continue until it passes continuous testing and goes on to continuous deployment.
And continuous deployment basically checks everything in, checks the configuration in, does the staging of the various live system, make sure there's nothing erroneous in there or something that's a vulnerability in terms of security issues, things like that and then it goes in deployment into operations and then we have operations, that may be a team of people that does that and they work directly with development and then continuous improvement which is a concept, in other words, it's typically not people but it's a concept that anything that is found along the way or any way in which we can make the software better, we're able to do continuous improvement, we're empowered to do it and then we can go back to the particular person, back to the automated tool and make the changes needed to make that software better.
So it's a much better system than the traditional way of doing things.
- Microservices and containers
- Complex, disturbed, serverless, and composite architectures
- DevOps integration
- High-performance solutions