Technology
GitHub's Infrastructure Crisis: How 3M New CPU Cores Weren't Enough
GitHub added massive compute capacity in 2024 but still faces scaling challenges as developer activity—driven by AI coding agents—surges beyond 2.9 billion commits per month.
The New Stack
GitHub now processes 2.9 billion commits monthly—more than double the 1.4 billion it struggled with just months earlier in April. The platform is also handling 130 million merged pull requests and launching 24 million new repositories each month. This explosive growth, fueled largely by the rise of AI-assisted coding agents, has exposed fundamental limits in the company's infrastructure even as it undertakes a major modernization effort.
The August Outage and Root Causes#
An eight-hour outage on August 17—GitHub's second major incident that month—exposed how quickly demand is overwhelming the platform's capacity. According to GitHub's postmortem, the problem wasn't faulty code or configuration errors but rather infrastructure hitting a breaking point.
When traffic spiked to unprecedented levels, a critical component in GitHub's Central US data center failed to scale in response. The strain rippled across the system, triggering authentication failures and service disruptions. CTO Vlad Fedorov acknowledged in the report: "We have made progress, but these incidents make clear that we must accelerate this work."
Migration to Azure and Infrastructure Limits#
Despite adding 3 million CPU cores and 120 petabytes of high-speed storage this year, GitHub remains stretched thin. The company has made progress migrating workloads to Azure—now handling 58% of platform load (up from just 12% in May) and half of all Git operations—yet growth outpaces available capacity.
GitHub's legacy data centers are now at maximum capacity, with all available power consumed by existing hardware. The company is prioritizing the shift to Azure's cloud infrastructure, but the transition isn't fast enough to keep pace with AI-driven development velocity.
Operational Changes Ahead#
Beyond infrastructure expansion, GitHub is restructuring how its systems operate. The company is strengthening testing procedures, improving monitoring and alerting, and implementing safer deployment practices. It's also isolating critical services to reduce the blast radius of future failures.
New retry limits, budgeted across service requests, and revised timeout settings are being deployed to prevent cascading failures when one component gets overwhelmed.
