GitHub rebuilds Git storage as agent traffic strains its infrastructure

GitHub is redesigning its Git storage to handle rising traffic, including a surge linked to AI coding agents. The Register reports that early internal tests showed a 35-fold improvement in write performance, though GitHub has not given a migration timeline.

Watch Desk analysis

What happened

The new design replaces Spokes, GitHub’s existing system, which stores repository copies across local disks and waits for a quorum of replicas to confirm each write. That can make a push wait on the slowest required replica.

Instead, the redesign writes each commit to Azure Blob Storage, which handles replication, while separate lightweight workers serve reads. Compaction and garbage collection move off the serving path and run in the background. The Register reports that the design aims to leave developer workflows, reviews and security controls unchanged.

The reported traffic figures put the rebuild in context: GitHub activity more than doubled between September 2025 and August 2026, reaching 473.3 billion events a month, and September 2026 saw 7.38 billion commits, five times the September 2025 figure. The Register also cites ten service incidents in April and nine in May that degraded performance.

Why it matters

The change targets a real infrastructure trade-off: preserving reliable, replicated storage while keeping writes moving as repositories and automated workloads multiply. If the new design works at scale, developers could see fewer slowdowns without having to change how they use GitHub.

But the 35-fold figure is from early internal tests, not a promise of a 35-fold improvement for every user. The practical test will be whether GitHub can migrate safely and keep performance steady under sustained traffic.

Our read

This is consequential plumbing, not a new button, and that is precisely why it matters. Agent-driven coding is putting pressure on the infrastructure beneath the tools; GitHub’s answer is to change how the storage works rather than ask developers to work around it. The architecture sounds promising. Production performance will get the final vote.

What to watch

• When GitHub plans to migrate repositories to the new architecture.
• Whether the internal write-performance gains hold under production workloads.
• Whether service reliability improves as traffic continues to grow.

Discussion spark: If AI coding agents keep driving up repository traffic, should GitHub prioritise a major storage redesign even if migration carries risk, or focus first on making today’s system more reliable?

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.