From the course: Architecting Big Data Applications: Batch Mode Application Engineering

Unlock the full course today

Join today to access over 24,700 courses taught by industry experts.

Product recommendations: Review the final architecture

Product recommendations: Review the final architecture

From the course: Architecting Big Data Applications: Batch Mode Application Engineering

Start my 1-month free trial Buy for my team

Product recommendations: Review the final architecture

“

- [Instructor] Let's now review the final architecture for the product recommendations use case. So here is the final architecture with all the technologies identified in the workflow. We have selected Apache Spark for the processing jobs, MongoDB for the transactions history database and Cassandra for the recommendations database. How does this architecture stack up? We use multiple database types in this architecture. In general, we want to reduce the number of database types in a single architecture, but in this case, we need them to support specific requirements. The processing jobs can leverage the querying and job distribution capabilities available in the Kafka and RDBMS inputs. This helps improve scalability of the pipeline. We are using a separate APA based product recommendation service. It's also possible to embed the product recommendation service inside the Spark recommendations job. The key decision point is…

Contents