Agile AI Platform Architecture with the Agile Cloud Manager
Part 9 of 10: AWS Example Appliance
The entire transcript of this video is given below the video so that you can read and consume it at your own pace. We recommend that you both read and watch to make it easier to more completely grasp the material.
An A.W.S. working example of an A.I. platform appliance is also included at the AgileCloudInstitute.io web site.
The A.W.S. example also includes a core example system that is composed of items that are shared by the other systems.
The contents of the systems are based on the managed services that A.W.S. provides, which are different than what other clouds provide.
The foundation of the core system includes shared elements, including:
- Networking.
- I.A.M. Roles.
- Storage.
- And other things.
Several services are defined in the core system in this example. Some of the services include:
- A service for managing users.
- Services for managing databases.
- Services for creating and assigning tags.
- Services for managing permissions.
There is also a system for managing data engineer resources.
A foundation in this example data engineer system includes definitions of I.A.M. roles, and configuration for an E.M.R. big compute cluster.
The data engineer system in this example also includes a service that manages an E.M.R. big compute cluster.
A data scientist system is also included in this example.
A foundation in the data scientist system includes storage that is shared by some of the services.
Several services in the data scientist example system include:
- A service that manages an E.M.R. big data compute cluster.
- A service that manages a Sage Maker domain.
- And other services.
The way that you could begin creating your own unique A.I. project management templates starting from this working example would be for you to start building out the templates associated with the Sage Maker domain service.
You might create one Sage Maker domain for each A.I. project. Therefore, after you define a well-spelled-out definition of a group of services related to each Sage Maker domain, you will be able to automatically deploy a full-featured, governable project management workspace for each new A.I. project simply by running one simple Agile Cloud Manager C.L.I. command.
This working example also includes preprocessors and postprocessors that do things that the templates cannot do. For example, the preprocessors and postprocessors in this working example do things like:
- Network administrative tasks.
- Copying and deleting data.
- Running automation.
- Removing things that templates cannot remove due to protection rules built into the cloud providers’ template processing rules.
If you take a look at this entire slide, you can start to see that you can rewrite every aspect of this working example to fit the unique needs of your organization. You might start by studying the working example before you begin iterating it.
The architecture section of the AgileCloudInstitute.io web site contains videos that describe this working example in some detail.
The marketplace section of the AgileCloudInstitute.io web site includes hands-on training that you can use to get this working example up and running right away.
Proceed to Part Ten: Next Steps
Back to Series Table Of Contents: Agile AI Platform Architecture With Agile Cloud Manager