Building and Operating Online Shops under Extreme Load
summary
Describes building and operating high‑traffic online shops that handled spikes of ~40,000 concurrent visitors, ~10,000 carts, and ~2,000 orders/minute. Uses CDNs to reduce origin traffic (~85%), heavy PHP profiling and framework choice for ~10× speedups, sharding with master–slave replicas plus an async, message‑queue‑backed write pipeline and inventory atomic counters to prevent oversell. Covers a caching error that caused ~4,000 wrong prices and large support work, and practical fixes like custom queue buffering and multiple payment service providers.
AI-written summary · about
chapters (20)
- 0:00Extreme load motivation
- 4:37Network DB load issues
- 7:04CDN file size optimization
- 9:10DB index optimization
- 10:56DB cluster scaling
- 13:24Profiling and load simulation
- 16:16Code optimizations
- 17:40Complex pricing challenges
- 20:47Caching dynamic content
- 23:56DB design and repositories
- 27:42Frontend caching performance
- 30:09System architecture config
- 32:42Distributed systems content
- 34:06Backend message queues
- 38:16Node.js performance
- 41:43Load testing tools
- 43:12Payment provider load balancing
- 46:47Async payment processing
- 48:15Atomic ticket counts
- 50:00Summary
related talks
35mSurviving Black Friday: Tales from an e-Commerce Engineer
18mBuilding Your Shopify App Backend for High-Volume Merchants
28mMastering Traffic Spikes with AWS
45mSurviving a Prime Time TV Commercial
37mScaling Rails for Black Friday and Cyber Monday
18mHigh-Load Systems: Challenges in Social Network Development
47mFrom Legacy to High-Load: Evolution of a Web Application
56mLarge scale websites performance optimisation tricks