Remote-First-Company – June 23, 2026 – VAST Data, the AI Operating System company, today announced that the German Climate Computing Center, Deutsches Klimarechenzentrum (DKRZ), has deployed the VAST AI Operating System to modernize a critical part of its high-performance computing environment for climate research.
DKRZ is Germany’s domain-specific HPC center for climate science, providing computing and data services to researchers across Germany and international climate research projects. The center manages approximately 240 petabytes of climate data, operates the World Data Center for Climate, and plays an important role in the global climate research ecosystem through initiatives such as the Earth System Grid Federation.
As climate models become more complex and AI begins to reshape scientific workflows, research computing centers are under pressure to support more users, more data and more demanding development environments. For DKRZ, that challenge extended beyond the largest simulation outputs. Researchers also needed reliable access to the everyday workspace where science takes shape: home directories, software stacks, Python environments, shared project spaces, code repositories and visualization scripts.
DKRZ selected VAST to replace and modernize its HPC home and software environment, giving researchers a more reliable foundation for daily work while providing infrastructure teams a simpler and more efficient platform to manage.
“Climate science depends on more than raw compute power. Researchers need reliable access to the environments where they develop models, build software, prepare workflows and analyze results,” said Michael Boettinger, Head Visualization and Outreach at DKRZ. “With VAST, we can provide a modern home and software environment that gives users more capacity, better reliability and fewer technical constraints in their daily work. The home system is central to researchers’ daily work. If it is unavailable, people may still have compute resources, but they cannot easily use the tools, scripts and environments they rely on. Keeping that environment available is a major improvement.”
For DKRZ, the decision to deploy VAST was driven by the complete platform package: snapshots, data reduction, monitoring, ease of administration, non-disruptive upgrades and efficient support for small-file-heavy workflows. These capabilities are especially important in research environments where users often create many similar but not identical software and Python environments, generating large numbers of small files that can become inefficient to manage on traditional systems.
“We were not looking for performance alone. We needed a system fit for the future too, with the right combination of features, reliability and administrative simplicity,” said Anna Fuchs, HPC I/O and Storage Engineer at DKRZ. “Snapshots, monitoring, compression and data reduction were all important, but the real value is that VAST brings these capabilities together in one platform. It allows our users to focus on their work instead of thinking about technical limitations and potential downtime.”
DKRZ has already seen strong results from VAST’s data reduction capabilities. Across production user environments, the center has observed a general reduction factor of approximately 5.5x, with specific software development and build environments achieving reduction factors of approximately 20x or more. This allows DKRZ to provide more usable capacity for researcher workflows without simply adding more hardware.
The platform has also enabled DKRZ to introduce shared environments between individual home directories and global software stacks, allowing project teams to maintain common software and Python environments without duplicating similar files across users.
VAST AI OS enables DKRZ to provide researchers with:
Uninterrupted Access to Critical Research Environments: Through non-disruptive upgrades and resilient system design, VAST with the VAST DataSpace keeps home directories, scripts and software environments continuously available.
Faster, More Responsive Development Workflows: VAST’s high-IOPS architecture efficiently handles Python environments, build systems and other small-file-heavy workloads.
More Usable Capacity: Built-in compression, deduplication and similarity-based reduction increase effective capacity without requiring equivalent physical expansion.
Fewer Maintenance Disruptions: Rolling software updates allow DKRZ to maintain and upgrade the platform without routinely taking user environments offline.
A Simpler, Better-Managed Research Platform: Integrated monitoring, snapshots and catalog capabilities give administrators greater visibility and control with less manual effort.
DKRZ is also exploring additional use cases for VAST, including internal services and code-related workflows that require reliable access from researchers and collaborators. The center is evaluating further platform capabilities as it considers how VAST can support future research infrastructure needs.
“Research computing centers are being asked to support more users, more data, more complex software environments and increasingly AI-driven workflows, often with lean teams and finite budgets,” said Christopher Huggins, Managing Director of AI & HPC, EMEA, VAST Data. “DKRZ understands that modern research infrastructure depends on more than just compute, and the data layer is not just a back-end storage concern. It is the foundation that allows researchers to develop, collaborate and turn computation into scientific insight.”
As climate science enters a new era of higher-resolution modelling, larger datasets and AI-assisted research workflows, DKRZ is building a more reliable, efficient and flexible foundation for scientific computing with VAST, and plans to expand the footprint are underway.