Content-Aware Storage (CAS)

The IBM Fusion Content-Aware Storage service is engineered to enhance the value of customers’ AI applications, such as retrieval augmented generation (RAG) for faster time to insights, reduced cost, improved performance, enhanced security, and streamlined operations.

CAS provides turn-key processing of unstructured data for use in multi-modal RAG applications using IBM Docling Multimodal services and NVIDIA NeMo Retriever Extraction microservices. It supports automated extraction of text, tables, and charts from Enterprise documents and makes the extracted information available for use in RAG applications. NVIDIA GPUs are used to accelerate the data extraction processing. For more information on CAS-related frequently asked questions, see Content-Aware Storage FAQs.

CAS uses IBM Storage Scale to cache existing Enterprise data residing on S3 sources to optimize processing with the IBM Docling Multimodal services and NVIDIA NeMo Retriever Extraction microservices without having to manage multiple copies of data. CAS also supports automated processing of data residing in the IBM Storage Scale filesystem. Incremental change processing is leveraged to detect when data is modified in the IBM Storage Scale filesystem and automatically process the changed data for use in RAG applications.