GitHub now sees 2.9 billion commits a month — and it can’t keep up

According to recent data, GitHub is now seeing an astounding 2.9 billion commits per month, along with 130 million merged pull requests and 24 million new repositories. This surge in activity has posed a challenge for the platform, with GitHub's infrastructure reaching its capacity just 18 months ago - compared to 1.4 billion commits in April.

This rapid growth is largely attributed to the increasing popularity of coding agents and the rapid pace of software development, although it's also no secret that GitHub has struggled to scale its architecture fast enough to keep up with the demand.

GitHub's troubles were laid bare on August 17 when the site experienced an almost eight-hour outage. In a postmortem report, GitHub CTO Vlad Fedorov noted that while progress has been made, these incidents highlight the need for accelerated work. He stressed that the outage was due to scaling issues, not any code changes, with the platform running out of capacity and the failure to scale a critical infrastructure component in their Central US data center.

Despite adding 3 million CPU cores and 120 petabytes of high-speed storage to its systems this year, GitHub's data center has maxed out. Fedorov explained that the team has installed as much hardware as allowed by power constraints while accelerating their migration to Azure, which now handles 58% of GitHub's platform load, up from 12% in May. Half of all Git operations are now served by Azure, indicating the company's shift towards the cloud provider.

While GitHub acknowledges the challenges posed by the increasing complexity and pace of change in the software development world, the company has also taken steps to improve its operational practices. These efforts include redirecting teams and resources towards availability, investing in stronger testing, safer rollouts, better observability, and more effective alerting.

The company is also working to isolate critical systems and remove shared dependencies between them to reduce the likelihood and impact of future outages.

In light of the recent outages, GitHub is implementing a few concrete changes, including applying consistent retry limits and budgets, and adjusting timeouts across service-to-service interactions to prevent retry storms and cascading load. However, GitHub's central role in the developer ecosystem, particularly in the open source sphere, creates opportunities for competitors.

Startups like Entire, founded by former GitHub CEO Thomas Dohmke, are betting on distributed systems to better handle agent-driven development workloads, with Origin by Cursor being one such example.