Главная
Study mode:
on
1
Intro
2
Niantic Mission
3
Niantic Real World Platform
4
Issues to solve
5
Service-gateway overview
6
Service mesh implementation Why not istio and others?
7
Service-gateway Next steps
8
Deployment overview
9
Goals
10
Scaling is a 4 steps process For example, scaling up
11
XDS to the rescue!
12
Current deployment
13
(very) High level plan
14
Preparing production environment
15
IP Deny List
16
Production migration - day 1+
17
Envoy live in Production
18
503 UH: UH (no healthy upstream hosts)
19
Back to load testing
20
Routing and DNS
21
Learnings
22
Extending Envoy: Niantic Frontend
23
Why Niantic chose Envoy
Description:
Discover how Niantic, creators of Pokémon GO, transitioned from NGINX to Envoy to handle millions of concurrent players worldwide. Learn about the challenges faced, the implementation of a service mesh, and the extension of Envoy to support proprietary protocols. Explore the scaling process using xDS, load testing strategies, and the final launch consolidation. Gain insights into Niantic's mission, their Real World Platform, and the reasons behind choosing Envoy. Delve into deployment overviews, production migration steps, and lessons learned from extending Envoy for Niantic's frontend.

How Niantic Switched Pokémon GO to Use Envoy

CNCF [Cloud Native Computing Foundation]
Add to list
0:00 / 0:00