6 min read
|
Saved February 14, 2026
|
Copied!
Do you care about this?
This article details Shopify's extensive preparations for the Black Friday Cyber Monday (BFCM) weekend, including year-round capacity planning and chaos engineering exercises. It outlines how the team conducts load testing and scale tests to identify and fix potential bottlenecks before peak traffic. The focus is on ensuring the infrastructure can handle unprecedented demand in 2025.
If you do, here's more
Shopify is ramping up its infrastructure for Black Friday Cyber Monday (BFCM) 2025, focusing on resilience and performance. The company began preparations in March, conducting bimonthly fire drills that simulated 150% of the previous year's traffic. In 2024, Shopify set new records, handling 57.3 petabytes of data, 10.5 trillion database queries, and peaking at 284 million requests per minute on edge servers. This consistent growth means that the company treats what once were peak periods as regular days, necessitating a complete overhaul of their BFCM readiness approach.
The preparation consists of three main components: capacity planning, infrastructure roadmap, and risk assessments. Traffic patterns are modeled using historical data, and the tech stack undergoes constant evaluation for necessary upgrades. Game Days, or chaos engineering exercises, allow Shopify to simulate failures and test critical user pathways like checkout and payment processing. These rigorous tests have built operational resilience and resulted in a Resiliency Matrix that documents vulnerabilities, recovery procedures, and incident response plans.
Load testing is another key strategy, using tools like Genghis to simulate user behavior and traffic bursts. This year, Shopify also faced unique challenges with a portion of their infrastructure that hadn't experienced holiday traffic before. Newly rebuilt analytics and ETL pipelines faced scrutiny through controlled experiments to ensure they could handle peak loads. Five major scale tests ran from April to October, with the final one reaching 200 million requests per minute, confirming that systems could withstand the anticipated surge.
Questions about this article
No questions yet.