ETL Pipeline to Achieve Reliability at Scale

Explore the design and implementation of a scalable ETL pipeline for handling high-volume financial transactions in an online betting exchange. Learn how to achieve reliability and accuracy in daily accounting reports while addressing challenges such as fault tolerance, fast data retrieval, and efficient computations. Discover the motivations behind the chosen tech stack, including Python3, Luigi, and Spark, and gain insights into solving key technical problems such as identifying and rerunning faulty steps, optimizing input/output operations, and enhancing computational speed. Delve into Spark's key concepts, execution processes, and integration with Luigi, as well as running Spark jobs on Amazon EMR for improved performance and scalability.

Intro
Accounting at markets
Fault tolerance and reliability
Efficient storage
Good performance
Spark key concepts
Execution on Spark
Spark job from Luigi
Spark on EMR
Shutdown EMR cluster