Organizations today are retooling their IT operations to embrace artificial intelligence (AI), machine learning (ML), and advanced analytics. But they’re challenged by data and workloads that are scattered around the globe, by the increasing time needed by AI training and inferencing workloads, and by the cost and scarcity of resources, especially graphic processing units (GPUs).
To address these challenges, IBM Storage Scale provides software-defined file and object storage for both structured and unstructured data, and IBM Storage Scale System is a hardware implementation of Storage Scale software that is optimized for the most demanding AI, high-performance computing, analytics, and hybrid cloud workloads.
To unlock the full potential of AI and ensure that fast GPUs aren’t being starved by slow IO, Storage Scale System 6000 supports the NVIDIA GPUdirect Storage protocol, which enables a direct data path between GPU memory and local or remote storage, such as NVMe or NVMe over Fabric (NVMe-oF). This GPUDirect architecture removes the host server CPU and DRAM from the data path, so the IO path between storage and the GPU is shorter and faster.
Discover how to enhance performance with NVIDIA DGX
Learn why the target use cases for parallel file systems extend to enterprise AI and analytics
Build a sustainable, low-energy and high-performance infrastructure, that is flexible for multiple concurrent workloads.
IBM Storage Scale System is designed to build an information supply chain for AI and the flexibility to access multiple data sources with parallel high performance platform.
Create an open hybrid cloud that modernizes infrastructure with global data unification from edge to cloud.
Solve AI data challenges with innovation, simplicity and business results.
Leverage an all-flash or hybrid scalable storage building block designed to create high-performing clusters for IBM Storage Scale in a sustainable IT architecture.
Take advantage of enterprise AI container-native storage solutions for OpenShift®.
Modernize your storage infrastructure for artificial intelligence (AI) and big data.
Find out how IBM and NVIDIA provide an end-to-end reference architecture to help you get faster AI results.
Get high performance data pipelines with cloud storage or other S3 data connected to IBM Storage Scale.
Global Data Platform, engineered to accelerate AI innovation and business growth with faster access to more data.