Optimise GitLab Object Storage for Scale and Performance

Streamline GitLab Object Storage for Growth and Efficiency

As your DevOps operations scale, efficient management of data storage becomes paramount. GitLab has recently introduced updated guidance and best practices to optimise object storage—whether you’re self-managing or operating at enterprise level. These insights are crucial for reducing storage-related bottlenecks and maintaining consistent performance.

GitLab uses object storage to house various forms of binary data including LFS files, CI/CD artefacts, uploads, and more. By employing object storage (e.g., AWS S3, Google Cloud Storage), GitLab instances gain scalability and increased reliability over traditional file systems for these assets.

However, operating at scale introduces several challenges: from access latency and request throttling by providers, to inefficient paths for large repositories. To mitigate this, GitLab provides targeted recommendations:

• Use a separate bucket per object type instead of sharing one across functions. This helps isolate request rates and simplifies access policies.
• Optimise bucket lifecycle policies to automatically remove expired data, such as expired artefacts or job traces, minimising waste and reducing costs.
• Take advantage of pre-signed URLs whenever possible to reduce traffic through GitLab Rails nodes, enhancing throughput and reducing server load.
• Monitor object storage metrics and configure alerts for early detection of latency, request failures, or quota issues.

GitLab is also rolling out performance improvements, such as defaulting async upload and storage background jobs, which ensure a smoother user experience without blocking the main application threads.

At IDEA GitLab Solutions, a GitLab Select Partner, we offer expert consulting and licensing services to businesses across the UK, Czech Republic, Slovakia, Slovenia, Croatia, Serbia, North Macedonia, and beyond. Our specialists can help you optimise your GitLab deployments, configure high-scale object storage, and build a DevSecOps-first culture tailored to your region’s unique needs. Contact us today to ensure your GitLab infrastructure is built for the future.

• What's New in Git 2.51.0 – Fast Status, Better Merges and Smarter Error Messages
• GitLab 17.11.7 Released: Key Fixes and Performance Enhancements
• GitLab’s Healthy Backlog Initiative: Streamlining Innovation
• Optimising Git Fetch Performance with Git Bundle URIs in GitLab
• How GitLab Reduced Repository Backup Times from 48 Hours to 41 Minutes