Deep Dive on Real World Use Case in Capitalone on Serverless and How It Enables Its Partner's...
summary
Capital One migrates a partner redemption pipeline from many EMR Spark jobs to a reusable serverless system that uses EMR Spark batch replaced by JSON 3DS configs to drive data source, data select/enrich and data sync, pulls from DynamoDB/S3, writes to staging and the data lake, and routes real-time flows through Kinesis + Lambda serverless pipeline with batch DB inserts and a dead-letter retry lambda for restartability and at-least-once guarantees; deployed to one major retailer then scaled to multiple partners.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:28About Capitalone
- 3:41About Me
- 4:59Agenda
- 8:15Partner Retention
- 9:49Partner Redemption
- 11:23Partner Configuration
- 12:39Partner Job
- 16:38Partner Data Sync
- 18:45Data Publishing
related talks
22mScaling to billions of requests the serverless way at Capital One
01h 01mServerless at scale
31mAccelerate business innovation with serverless applications
34mUnleashing Agility: How Capital One Embraced COTS for NextG...
39mBuilding Highly Scalable Microservices with Serverless
54mTableau Server on AWS enables analytics agility for Capital One
34mDesigning Data Intensive Applications In Serverless Architecture
42mEnterprise Serverless Adoption. An Experience Report