Lift and shift is a common enterprise strategy for moving into the cloud. In this video, take a closer look at important considerations for this scenario.
- [Instructor] As we move through this course, we're now going to work with scenarios and services. Because the course would be super long if I covered every aspect of every scenario, what I've done is I've picked common scenarios and I'm pulling out parts of them so that I can showcase best use of services, both existing services, where I've seen capabilities maybe not used or not used correctly, and also new services, since there are so many new services and new service capabilities that are interesting to enterprise. So it's basically highlights out of the architectural scenarios. I'm also going to show you how in my work as a cloud architect, I use GCP resources. GCP has not been known for a strength of their resource offering, and one of the things I've learned to navigate in the years of working with the Google Cloud is how to read between the lines in their documentation. How to understand how to make it work. How to work with code and examples. So I'm going to basically show you how I would work on parts of architectures going throughout the rest of this course. So the first one is we're going to talk about lift-and-shift. So this is a very common enterprise strategy going into the cloud. The idea is you have some sort of application and I just pulled a standard architecture for a very classic web server. This could be for a customer facing web app or it could be for a company focused internal application. Really doesn't matter at this point. But the idea is that we need to understand what are the parts and pieces and how we would lift-and-shift them into GCP using best practices. So you can see we've got an HTTP server, got some firewalls, a web app server, a database, and a business logic server. Of course, this is somewhat oversimplified, but it does give us a reference point for starting. To think about lift-and-shift, if we think about all the parts and pieces, we've got compute, files, data, and other. So in the world of compute, we're going to start with the classic virtual machines. Now in other architectures and actually at the end of this section, we're going to look at containers. In subsequent architectures, we're going to look at functions. But right away I would say that going directly to functions kind of is too big of a leap for a lift-and-shift kind of scenario. So really on GCP, we're now looking at either starting with migrating VMs or there is a new service that allows us to migrate VMs as containers. From a file standpoint, we would usually start with Google Cloud storage. However, there is a newly released service that we could consider here as well, which we'll talk about throughout the scenarios. For data, typically I recommend customers that are using relational databases start with Cloud SQL. It is interesting that having a managed relational database service has come late into GCP, and this is because Google themselves use relational database stores but they also use other type of NoSQL stores. And those NoSQL stores are used for their own massive datasets. So it's kind of an afterthought or has been in the GCP ecosystem. However, one of the improvements over the past 12 months or so has been improvement in the offering quality of their Cloud SQL or their managed MySQL or PostgreSQL databases. Also, they announced last week that they will be supporting SQL Server and Cloud SQL, which gives more options for lift-and-shift. In previous movies, we talked about infrastructure considerations. So I won't repeat that information, but it goes without saying that we'd have to have VPC interconnect and IAM objects set up correctly for our particular scenario. Also, as we think about lift-and-shift, if you want to go deeper into this architecture and best practices, there is a white paper that I've referenced here that literally gives you very prescriptive guidance on how to lift-and-shift a line of business application onto the Google Cloud platform. Like I said, in this course, we're going to focus on aspects of lift-and-shift rather than the entire end-to-end movement process. But I realize some of you will have this use case. So I would recommend this white paper if this is something you're actually going to attempt as well.
- Enterprise concerns
- Enterprise scenarios
- Setting up your organization’s account
- Managing billing
- Enterprise compute services
- Enterprise storage and database services
- Enterprise data pipelines
- GCP developer and DevOps tools