Professional Data EngineerDesigning data processing systemsEasy

A global media company needs to process millions of user interaction events per second, such as clicks, views, and searches, to power real-time personalization and analytics dashboards. The data needs to be ingested reliably and cost-effectively, with the ability to scale dynamically to handle peak loads. Which Google Cloud product should the company use for ingesting these events?

  1. ABigQuery
  2. BCloud Storage
  3. CCloud Pub/Sub
  4. DCloud SQL
Show answer & explanation

Correct answer: C. Cloud Pub/Sub

Cloud Pub/Sub is designed for highly scalable and reliable real-time messaging, making it ideal for ingesting millions of events per second. It decouples producers and consumers, allowing for asynchronous processing and dynamic scaling.

Why the other options are wrong

  • A. BigQuery is a data warehouse for analytics, not an ingestion service for real-time streams.
  • B. Cloud Storage is primarily for object storage, not real-time event ingestion and messaging.
  • D. Cloud SQL is a relational database and not suitable for ingesting millions of events per second at scale.

Cloud Pub/Sub for Event Ingestion

Cloud Pub/Sub is a real-time messaging service designed for ingesting large volumes of events from various sources and delivering them to multiple subscribers.

  • Scalable to millions of events per second.
  • Decouples publishers and subscribers.
  • Provides reliable, low-latency message delivery.

Memory trick: Pub/Sub handles the 'publish' of data and 'subscribe' to its flow, like a real-time news feed.

More Designing data processing systems questions