BeeGFS High-Performance Storage Software, Sovereign by Design

BeeGFS delivers high-performance parallel storage built for organizations that demand both extreme performance and uncompromising control.
Designed around the principle of sovereignty across five critical dimensions: Data, Technology, Hardware, Operations, and Compliance. BeeGFS empowers you to scale without dependency, innovate without restriction, and govern without compromise. 

Sovereign across five dimensions: 

Data
Technology
Hardware
Operations
Compliance

Data Sovereign:

  • You own your data, where it lives, how it moves, and who can access it. Always.
  • No lock-in.

Technology Sovereign:

  • Deploy anywhere: on-prem, private cloud, sovereign cloud, hybrid.
  • Seamless compatibility with modern kernels and software ecosystems, even alongside competing storage stacks.

Hardware Sovereign:

  • Your hardware. Your choice. BeeGFS runs on any hardware from any vendor
  • Deployable across any storage tier, from Flash storage up to Tape 

Operations Sovereign:

  • Easy to deploy, easy to use, easy to maintain, easy to scale
  • Zero  downtime

Compliance Sovereign:

  • Source available software.
  • Developed in Europe and deployable entirely under your governance

Whether deployed on-premises, in private or sovereign clouds, or across hybrid environments, BeeGFS ensures your infrastructure remains fully yours; transparent, flexible, and engineered for independent growth from edge to exascale.

Why BeeGFS?

Because performance should never come with complexity or compromise.

BeeGFS is a European-governed, high-performance parallel file system, architected for large-scale HPC and AI clusters, deployable entirely under your control, and trusted by thousands of organizations worldwide. Proven across data-intensive industries including life sciences, energy, defense, and academic research, BeeGFS supports environments ranging from small clusters to the world’s largest supercomputers.

With commercial licensing and enterprise support available, BeeGFS provides the stability and operational confidence required for mission-critical deployments. The result? Maximum throughput, reduced bottlenecks, and the freedom to scale without rearchitecting your environment.

Explore the BeeGFS

Hive of Possibilities!

ThinkParQ offers BeeGFS as a self-supported Community Edition and a fully supported Enterprise Edition with additional features and functionalities

BeeGFS

Community Edition*
What’s included

Everything to get you started with the community version of the BeeGFS High Performance Scratch File System that is trusted by our global community.

Capacity-based threshold apply, need more?. Contact our sales team here!

BeeGFS stores Access Control Lists (ACLs) as Extended Attributes of metadata files on the metadata server.

BeeGFS now offers systemwide monitoring capabilities, allowing you to easily monitor the performance and health of your BeeGFS system. With the inclusion of Grafana dashboards, you can visualize and analyze the collected data to gain valuable insights into the system’s behavior and optimize its performance.

Receive all our in-depth information, valuable insights, and the latest updates directly from us.

Up to 5 Compute nodes contributing storage

Enterprise Edition 

BeeGFS

Hive Enterprise
What You’ll Get 

The BeeGFS File System plus a Commercial license that includes professional support along with the following features:

The BeeGFS Community edition is capped at 1,200 TB total, with a limit of 1,000 TB for HDD-based storage and 200 TB for SSD-based storage (NVMe, SATA, or SAS). Enterprise lifts those limits: your license covers the capacity you provision, so storage grows with your infrastructure instead of stopping at a fixed line.

To learn more, contact our sales team here!

BeeGFS stores Access Control Lists (ACLs) as Extended Attributes of metadata files on the metadata server.

BeeGFS now offers systemwide monitoring capabilities, allowing you to easily monitor the performance and health of your BeeGFS system. With the inclusion of Grafana dashboards, you can visualize and analyze the collected data to gain valuable insights into the system’s behavior and optimize its performance.

Receive all our in-depth information, valuable insights, and the latest updates directly from us.

Up to 5 Compute nodes contributing storage

Integrate High Availability features into your infrastructure.

BeeGFS provides support for metadata and file contents mirroring. Mirroring capabilities are integrated into the normal BeeGFS services, so that no separate services or third-party tools are needed.

Quota enforcement allows the definition and application of quota limits in the whole system.

Storage targets with different performance characteristics (NVMe QLC, NVMe TLC, or HDD) can be assigned in BeeGFS Storage Pools. Data can be transparently tiered between different storage pools within the same namespace, using a policy or workflow automation tools (auto-tiering).

RST is a built-in feature for bulk data movement between BeeGFS and other file or S3-compatible storage systems, whether on-premises or in the cloud, enabling simple data staging (tiering) and ingress/egress with minimal changes to existing application workflows.

Parallel data mover between BeeGFS and other POSIX based file systems to load / offload data in significantly less time

A dynamic event listener API that supercharges BeeGFS indexing without sacrificing metadata performance, allowing a much tighter integration in HPC and AI workflows.

Index is a hierarchical index that stores file system metadata to enable users to run queries and searches on a file system with millions to billions of files and receive results in seconds without affecting the performance and efficiency of the file system itself.

Migrates file contents efficiently between storage targets in the background. Metadata services coordinate with storage nodes to move chunks directly between targets and update stripe patterns without needing to rewrite files completely. Common use cases include redistributing data across new targets when expanding the file system, or moving infrequently accessed data from hot to cold storage pools.

Why AI and HPC Teams choose BeeGFS?

Its' Fast

Throughput scales near-linearly with hardware – from GB/s on a single rack to TB/s at supercomputer scale.

It's Simple

Deployed in minutes. Zero-downtime upgrades, maintenance, and scaling – no specialist team required.

It's Sovereign

Software-defined. Your hardware, your network, your cloud, POSIX everywhere. No lock-in, no re-architecting, no external dependencies.

It’s Current.

New Linux kernels and distributions supported in weeks, not quarters.

It’s Flexible.

Move cold data to cheap storage. Source your own hardware. Redeploy existing hardware.

It’s Proven.

Trusted in production AI & HPC clusters and Top 500 class supercomputers – for over a decade.

How BeeGFS Works.

BeeGFS is engineered for independent scaling of metadata performance, storage capacity, and data throughput – growing exactly where workload demands increase, without re-architecting applications.

It runs anywhere Linux runs: on-premises, in any cloud, or on lightweight ARM systems like a Raspberry Pi, with packages for all major distributions. New services can be added without downtime. From a user’s perspective, BeeGFS behaves like any shared Linux file system, with standard POSIX semantics and a global namespace; the client is a Linux kernel module integrated directly with the VFS layer for a low-overhead data path.

BeeGFS supports heterogeneous storage tiers (NVMe, SSD, HDD) within a single namespace, with transparent, policy-driven data migration between them, and file contents can be offloaded to S3-compatible storage while keeping a consistent namespace.

Operations are optimized for highly concurrent parallel workloads: files are striped across storage nodes by default, so clients share the cluster’s aggregate bandwidth. RDMA support (InfiniBand, OmniPath, RoCE) and NVIDIA GPUDirect Storage integration reduce latency and CPU overhead for AI training and inference. Built-in bulk data movement tools support typical HPC and AI staging and ingress/egress workflows with minimal changes to existing applications.

From a single cluster to exascale, BeeGFS scales without requiring you to rearchitect at every step.

To learn more about the latest version of BeeGFS or to look deeper into our documentation, click here!

FAQ’s

It is easy to contact our sales team. Please send a direct email to sales@thinkparq.com

No, BeeGFS is not open source however source available.

BUG is the BeeGFS User Group meeting that takes place twice a year in Hamburg along ISC and is the USA alongside SC. You can learn more about BUG on the BeeGFS HUB Page here: https://www.beegfs.io/c/beegfs-hub/

Please visit our download page here: https://www.beegfs.io/c/download/

Learn more about the BeeGFS System  training and how to sign up for a training session here.

To access the Benchmarking tools please visit the BeeGFS Wiki here.

To access the BeeGFS documentation please click here.

To read the BeeGFS EULA, please click here.

ThinkParQ offers BeeGFS as a self-supported Community Edition (free) and a fully supported Enterprise Edition with additional features and functionalities. To learn more, please contact our sales team here!

The BeeGFS Container Storage Interface (CSI) driver provides high performing and scalable storage for workloads running in container orchestrators like Kubernetes. This driver allows containers to access existing datasets or request on-demand ephemeral or persistent high speed storage backed by BeeGFS parallel file systems.

The driver can be easily deployed using the provided Kubernetes manifests. Optionally the BeeGFS CSI Driver Operator can be used to automate day-1 (install/ configure) and day-2 (reconfigure/update) tasks for the driver. This especially simplifies discovery and installation from Operator Lifecycle Manger (OLM) enabled clusters. Multi-arch images supporting amd64 and arm64 Kubernetes nodes are provided for the BeeGFS CSI driver and operator.

See More!

BeeGFS is primarily C++ and C for its core components (metadata, storage, and client kernel module), with newer services, including the Management service and the command-line tool – being rewritten in Rust and Go respectively.

BeeGFS builds and runs on Linux, with documented build instructions for Red Hat/CentOS and Debian/Ubuntu family distributions.

HPC, AI, Life Sciences, Energy, Oil and Gas, Media and Entertainment, Space, Defense, Academic Research

A 60 day trial version of BeeGFs Enterprise is available here: https://www.beegfs.io/license/

RST is a built-in feature for bulk data movement between BeeGFS and other file or S3-compatible storage systems, whether on-premises or in the cloud, enabling simple data staging (tiering) and ingress/egress with minimal changes to existing application workflows.

Migrates file contents efficiently between storage targets in the background. Metadata services coordinate with storage nodes to move chunks directly between targets and update stripe patterns without needing to rewrite files completely. Common use cases include redistributing data across new targets when expanding the file system, or moving infrequently accessed data from hot to cold storage pools.

BeeGFs is developed in Europe by ThinkParQ GmbH and is designed around the principle of sovereignty across five critical dimensions: Data, Technology, Hardware, Operations, and Compliance. BeeGFS empowers you to scale without dependency, innovate without restriction, and govern without compromise.

Both editions run the same core BeeGFS file system — same performance, same POSIX behavior, same architecture. The difference is capacity and capability.

Community Edition is free and includes ACLs, Grafana-based monitoring, and BeeOND (with limits), with combined storage capped at 1,200 TB (1,000 TB HDD + 200 TB SSD).

Enterprise Edition (BeeGFS Hive) Enterprise lifts those limits of the Community Edition: your license covers the capacity you provision, so storage grows with your infrastructure instead of stopping at a fixed line. Enterprise adds features including High Availability, mirroring, quota enforcement, storage pools with auto-tiering, Remote Storage Targets, Copy, Watch, Index, and background data rebalancing. It also includes commercial support.

Not sure which fits? Start a 60-day trial or contact sales.

If you want high performance without the setup headache choose BeeGFS.
If you lack a dedicated file-system administration team but still need top-tier parallel performance, BeeGFS is highly recommended and used by many. It is hardware-independent, lightweight, and handles the random I/O common in AI data preparation incredibly well
Get Started with BeeGFS