How does Pokémon Go scale to millions of requests?

Опубликовано 16 октября 2021, 16:00

Pokémon Go → goo.gle/3mRIO09
Niantic Labs → goo.gle/3BDtRoB
Anatomy of a BigQuery Query → goo.gle/30ioOvU

Millions of people are playing Pokémon GO, but did you know it’s built on Google Cloud? In this episode of Architecting with Google Cloud, Priyanka Vergadia speaks with James Prompanya, the Technical Manager at Niantic Labs about Pokémon GO’s architecture and how it scales for a large number of players. Listen as they share how they use GKE and Spanner for scaling and how their data science team works with BigQuery, Dataflow, and Pub/Sub for the data analytics!

Chapters:
0:00 - Intro,
0:42 - What is Pokémon GO and GO Fest?
2:08 - How does Pokémon GO scale to handle peak traffic?
4:00 - Why does the team use Spanner and GKE?
5:18 - Request flow and architecture
7:58 - Data processing workflow with Pub/Sub, Dataflow and BigQuery
10:50 - Scaling data processing for 10-20GB of data per day
11:26 - How do they monitor the health of systems
12:11 - What’s next - Game Servers
12:29 - Summary

Architecting with Google Cloud → goo.gle/3g7xAC9
Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech

#ArchitectingWithGoogleCloud

product: Cloud - General; fullname: Priyanka Vergadia;