Menu
IBM Storage Scale System is IBM’s appliance-based deployment of IBM Storage Scale, designed to support AI, high-performance computing, advanced analytics and other data-intensive workloads. IBM positions the platform as a way to combine IBM Storage Scale software with integrated hardware so organisations can deploy a global data platform with high-performance access to data across distributed environments.
For organisations that need more than software alone, IBM Storage Scale System provides a hardware implementation of IBM Storage Scale that is built for performance, scalability and simpler deployment. IBM’s current product family includes the IBM Storage Scale System 3500 and IBM Storage Scale System 6000, each aimed at different performance and scale requirements.
IBM Storage Scale System is the appliance version of IBM Storage Scale. IBM describes it as being built on Storage Scale software, with the system family designed to accelerate AI, HPC and advanced analytics workloads by integrating IBM Storage Scale software with NVMe flash technology. IBM also presents the platform as part of a global data approach for AI and enterprise data.
This matters because it distinguishes Storage Scale System from Storage Scale software-only deployments. IBM Storage Scale is software-defined storage and a clustered file system, while Storage Scale System is the hardware implementation of that software for organisations that want an appliance-based route into the platform.
At the software layer, IBM Storage Scale is described by IBM as a clustered file system and a global data platform for AI, high-performance computing and advanced analytics. IBM highlights massively parallel file system performance, support for file and object data, and integration across data centres, cloud and edge environments. IBM Storage Scale System builds on that software foundation in an appliance form factor.
IBM Storage Scale is the software platform, while IBM Storage Scale System is the pre-integrated appliance family built around it. That distinction should stay visible throughout the page, but the primary focus here should remain the appliance family because that is the ranking topic.
IBM Storage Scale System 3500 which is 2U delivers enterprise-grade performance in a compact form factor. Its simplicity, flexibility and a modular building-block design, making it suitable for midsize organisations and edge deployments.
Compact, modular, enterprise-grade, and suited to organisations that need IBM Storage Scale System in a smaller footprint.
IBM positions the IBM Storage Scale System 6000 for extreme performance and scalability to support data-intensive workloads. IBM 6000 is NVIDIA-Certified Storage and delivers GPU-optimised throughput to support faster insights and decision-making.
IBM Storage Scale System 6000 as a hardware implementation of IBM Storage Scale software that is optimised for demanding AI, HPC, analytics and hybrid cloud workloads.
Performance at scale, parallel access to data, AI and analytics readiness, and access to file and object data across distributed environments. IBM also frames Storage Scale System as part of a global data platform approach for AI and enterprise data.
Organisations with fast-growing data volumes, AI pipelines, HPC workloads, or analytics environments that need an appliance-based implementation of IBM Storage Scale rather than a software-only approach.
IBM Storage Scale is the software platform. IBM describes it as software-defined storage and a clustered file system for AI, HPC and advanced analytics, with support for file and object data across data centre, cloud and edge environments.
IBM Storage Scale System is the appliance family built on that software. Integrated hardware implementation for organisations that want a more direct route to deploying IBM Storage Scale for enterprise, AI and high-performance workloads.
IBM connects Storage Scale and Storage Scale System to AI, HPC, analytics and global data platform use cases. IBM’s reference architecture and resource pages also tie Storage Scale System 6000 directly to NVIDIA DGX SuperPOD, NVIDIA DGX BasePOD and NVIDIA Cloud Partner architectures.
That gives you a solid, IBM-backed use-case set for the page:
