Metadata search
Metadata search index
Turned the open-source GUFI indexing engine into fast metadata search: native capture plugin, concurrency and crash-consistency fixes, the index CLI, and its release pipeline.
Engineering partner to ThinkParQ on BeeGFS
ParoScale is an engineering partner to ThinkParQ, the company behind BeeGFS. We work as an embedded engineering team, building features directly in the BeeGFS codebase, along with the kernel compatibility, CI/CD, test automation and monitoring needed to release each version. We don't offer BeeGFS installation or consulting on our own.
Features we built
High availability
Mirroring & resync fixes
Fixed metadata-server crashes and buddy-resync failures in mirrored deployments.
Performance
Faster metadata path
Optimised stat retrieval, eliminating unnecessary syscalls.
Tooling
CLI
Contributions to the Go CLI replacing the legacy C++ tool, including metadata-search commands.
The full lifecycle
Kernel compatibility
- Shim layer across VFS and memory-management API churn, through Linux 6.12
- Idmapped mount support
- Parallel kernel module: mount two major versions during a live upgrade
CI / CD
- 20 OS × arch release builds, native ARM runners
- Support matrix auto-discovered from upstream distro repos
- Separate functional, coverage, perf-regression and lint pipelines; GPG-signed packages
Test automation
- Testrunner v2: distributed scheduler across hosts and VMs
- Automated pytest and interactive CLI on shared primitives
- Self-tests for coredumps, log-watching and crashes; 9 VM targets plus a physical lab
Production monitoring
- 15+ pre-built Grafana dashboards
- Metadata, client ops by node and user, storage targets
- Installable alerting rules
“We have had the pleasure of working closely with Manisha and the team at Paroscale Technologies for the past few months, going straight into forward development with very little time to adopt the development environment, as well as the code base for our parallel file system BeeGFS. The meticulous work that Paroscale has already provided, exceeded our expectations. It is a pleasure to have Paroscale as a partner.”
Storage slower than your compute?
Send us your topology, workload and a benchmark run. We will get back to you.