From the course: Cloud-Based AI Solution Design Patterns

Unlock this course with a free trial

Join today to access over 24,600 courses taught by industry experts.

Hybrid data processing

Hybrid data processing

From the course: Cloud-Based AI Solution Design Patterns

Start my 1-month free trial Buy for my team

Hybrid data processing

“

- AI systems will often need the compute power and storage capacity offered by cloud providers. However, it isn't always the best idea to store all of an organization's data in a public cloud. There is sensitive data or critical data that is vital to the organization, or there can be data that is legally required to be stored in certain geographical locations. Sometimes the best option for this type of data is for it to be stored on premises within the organization itself. The hybrid data processing pattern addresses this challenge by providing an architecture that distributes an AI solutions data across both on-premises and cloud environments. If both the processing and the storage of some of the data needs to physically occur within the organization, then it may make sense for the AI system to also reside on premises alongside the data. It would then remotely access the remaining data that's located in a cloud. A potential downside of this approach is that the AI system would be…

Contents