GitHub currently processes 2.9 billion commits per month, according to a postmortem report published by the company following its nearly eight-hour outage on August 17. The platform also merges 130 million pull requests and adds 24 million new repositories monthly, figures disclosed as the company explained its second major service disruption in August alone. GitHub CTO Vlad Fedorov attributed the dramatic growth largely to the rising popularity of coding agents and the rapid pace of change in software development.
The surge represents a sharp jump from April, when GitHub was handling 1.4 billion commits and already struggling with capacity constraints. The company has migrated 58% of its platform load to Azure, up from just 12% in May, and now serves half of all Git operations through Microsoft's cloud infrastructure. GitHub added 3 million CPU cores and 120 petabytes of high-speed storage this year, while its own data centers have reached maximum capacity, with hardware installations limited only by available power.
The August 17 outage began when traffic hit a new peak and a critical infrastructure component in the company's Central US data center couldn't scale to meet demand, Fedorov explained. The resulting capacity pressure spread through GitHub's systems, triggering authentication failures and disrupting multiple services. "We have made progress, but these incidents make clear that we must accelerate this work," Fedorov stated in the postmortem. He emphasized that the outage stemmed from scaling issues rather than code changes, as the infrastructure simply ran out of capacity and those problems cascaded across the platform.
GitHub's operational practices haven't kept pace with the increasing speed and complexity of changes hitting the platform, according to Fedorov. The company has redirected teams and resources toward availability, investing in stronger testing, safer rollouts, better observability, and more effective alerting, though Fedorov acknowledged this work remains incomplete. GitHub is also isolating critical systems and removing shared dependencies between them to reduce the likelihood of outages and limit their impact when they occur. To address the immediate triggers of recent disruptions, the company is applying consistent retry limits and budgets while adjusting timeouts across service-to-service interactions to prevent retry storms and cascading load.
The platform's reliability struggles have created openings for competitors to launch rival products, including Entire, the startup founded by GitHub's former CEO Thomas Dohmke, which is betting on a distributed system better suited to agent-driven development workloads. Cursor is also entering the space with Origin. GitHub remains central to the developer and open source ecosystems, but its recent issues have exposed vulnerabilities in its architecture even as it races to migrate infrastructure and add capacity. The company's technical debt appears to have caught up with it at precisely the moment AI-powered coding tools are driving exponential growth in platform usage. For enterprises relying on GitHub as mission-critical infrastructure, the twin August outages raise questions about diversification strategies and the hidden costs of consolidation around a single platform, particularly as development workflows become more automated and less tolerant of downtime.