Story
Confluent
miHoYomiHoYo Uses Flink on Confluent for Event Streams
miHoYo, a communication services organization in China, uses Flink on Confluent from Confluent to support real-time operations for platform engineers.
Value results
| Category | Value result |
|---|---|
| Productivity | Handoffs in real-time operations sit in a shared queue instead of a mailbox trail |
| Risk and compliance | Flink on Confluent is the governed place platform engineers use for real-time operations |
| Capability | New joiners can see how real-time operations actually runs |
Story
Inside miHoYo, real-time operations used to depend on whoever still had the latest file. That pattern is common in communication services groups working out of China. Platform engineers needed a system that would still make sense after the original project team moved on.
miHoYo uses Flink on Confluent from Confluent as the working layer for event streams. Confluent is the data streaming platform built around Apache Kafka, used to move events between applications in real time. The practical change is simple: real-time operations has a home, and reviews happen there instead of in a forwarded thread.
Nothing in this writeup invents a savings number. What miHoYo gets from Confluent is a durable place to run real-time operations and a way for platform engineers to see the same event streams at the same time.
