Complete the code to identify the main component responsible for storing user watch history.
The component that stores user watch history is called the [1].The Watch History Store keeps track of videos watched by users, which is essential for personalized recommendations.
Complete the code to specify the type of database best suited for storing video metadata.
For storing video metadata like title, description, and tags, a [1] database is preferred.
Relational databases are ideal for structured data like video metadata with defined schema.
Fix the error in the description of the recommendation algorithm type.
The system uses a [1] filtering algorithm that recommends videos based on user similarity.Collaborative filtering recommends items based on similarities between users' preferences.
Fill both blanks to complete the request flow for serving a video recommendation.
User request -> [1] -> [2] -> Response with recommended videos
The Recommendation Service processes the request using data from the Video Metadata Store to generate recommendations.
Fill all three blanks to complete the capacity estimation formula for the recommendation system.
Total requests per second = [1] * [2] * [3]
The total requests depend on how many users are active, how often they request, and how many replicas handle the load.