Ads Recommendation System Design
Explore the design and scaling of ads recommendation systems focused on ad click prediction. Understand data estimation, system components like data lakes and feature stores, and methods to maintain low latency with high traffic. Learn strategies for model retraining and load balancing to optimize ad ranking performance in production.
4. Calculation and estimation
Assumptions
- 40K ad requests per second or 100 billion ad requests per month
- Each observation (record) has hundreds of features, and it takes 500 bytes to store.
Data size
- Data: historical ad click data includes [user, ads, click_or_not]. With an estimated 1% CTR, it has 1 billion clicked ads. We can start with 1 month of data for training and validation. Within a month we have, 100 *