How to Enable 100 Million Logins Without Downtime ...
summary
Betclic describes splitting a legacy monolith into domain microservices, adding caching and Kong gateway throttling, and migrating parts to AWS to handle massive load. Load and failure tests were executed in production using Gatling to generate traffic and monitored with AppDynamics and Datadog, which revealed DB pooling, SSL and network limits that were fixed; an internal portal coordinated scenarios. The result was roughly one million concurrent players and about 100 million logins in a month with 100% availability and no major incidents.
AI-written summary · about
chapters (8)
- 0:00Start
- 11:58Match evenings
- 19:10Cloud migration kickoff
- 26:33Types of load testing
- 30:30Examples of improvements
- 37:28A few limitations
- 38:23Next steps
- 40:47Technical blog
related talks
47mHow to Enable 100 Million Logins Without i…
26mHow We Prepped Our Infrastructure for a Super-Bowl Sized Traffic
54mGetting to 300 million requests a month with a one man AI/Dev-Ops team
46mManaging a Billion Transactions with Tech, Tactics, and Teamwork
23mPerformance and Scaling for Special Events at eTF1
28mReliability Lessons from Running High Traffic Services for More Than 10 Years
31mHow We Went From Two Major Outages to 99.98% Reliability in 6 Months
21mTransforming Legacy Banking Systems through Machine Learning