Leveraging the Nvidia AI Information Platform reference structure, these new options are the newest within the collaboration:
Augmenting Unstructured Information Processing: With IBM’s new content-aware storage (CAS) functionality, enterprises can extract insights from unstructured knowledge, with out compromising belief and security, to responsibly scale and improve AI purposes like retrieval-augmented era (RAG) and AI reasoning. IBM Storage Scale will reply to queries utilizing the extracted and augmented knowledge, dashing up communications between GPUs and storage. CAS can be embedded within the subsequent replace of IBM Fusion deliberate for the second quarter of this yr.
Extra Accessible AI: IBM plans to combine its watsonx choices with Nvidia’s to offer entry to main AI fashions throughout a number of cloud environments. It will permit organisations to leverage watsonx.ai, IBM’s enterprise-grade AI platform and developer studio, to develop and deploy AI fashions into purposes whereas utilising externally hosted fashions.
Help For Compute-Intensive Workloads: IBM Cloud has expanded its Nvidia accelerated computing portfolio by saying the provision of the H200 Tensor Core GPU situations on IBM Cloud. With its massive reminiscence capability and excessive bandwidth, the H200 situations are engineered to fulfill the calls for of contemporary AI workloads and bigger basis fashions.
Agentic AI: IBM Consulting is introducing AI Integration Companies to assist enterprises handle processes with agentic AI utilizing Nvidia Blueprints, comparable to industry-specific workflows that require agentic AI on the edge. Instance use instances embody autonomous inspection and upkeep within the manufacturing {industry} or proactive video knowledge evaluation and anomaly response within the vitality {industry}.
Optimising Workloads Throughout Hybrid Cloud: IBM Consulting helps enterprises construct and handle compute-intensive AI workloads throughout hybrid cloud environments leveraging RedHat OpenShift and Nvidia AI. This contains applied sciences like Nvidia AI Foundry, NeMo, AI Enterprise, Blueprints, and Clara to speed up high-compute, complicated duties, whereas managing AI governance, knowledge safety and compliance necessities.
