An "approximate nearest neighbour" search (ANNS) algorithm optimised for SSDs, KIOXIA’s AiSAQ software has been designed to deliver scalable performance for retrieval-augmented generation (RAG) without placing index data in DRAM - and instead searching directly on SSDs.
Generative AI systems require significant compute, memory, and storage resources, so while they have the potential to drive transformative breakthroughs across various industries, their deployment often comes with high costs.
RAG is a critical phase of AI that refines large language models (LLMs) with data specific to the company or application.
A central component of RAG is a vector database that accumulates and converts specific data into feature vectors in the database. RAG also utilises an ANNS algorithm, which identifies vectors that improve the model based on similarity between the accumulated and target vectors.
In order for RAG to be effective, it must rapidly retrieve the information most relevant to a query and traditionally, ANNS algorithms are deployed in DRAM to achieve the high-speed performance required for these searches.
KIOXIA’s AiSAQ technology provides a scalable and efficient ANNS solution for billion-scale datasets but with negligible memory usage and much faster index switching capabilities.
AiSAQ technology allows large-scale databases to operate without relying on limited DRAM resources, enhancing the performance of RAG systems and eliminates the need to load index data into DRAM, enabling the vector database to launch instantly. This supports seamless switching between user-specific or application-specific databases on the same server for efficient RAG service delivery.
The technology has also been optimised for cloud systems by storing indexes in disaggregated storage for sharing across multiple servers. This approach dynamically adjusts vector database search performance for specific users or applications and facilitates the rapid migration of search instances between physical servers.
"The KIOXIA AiSAQ solution paves the way for almost infinite scaling of RAG applications in Generative AI Systems based on flash-based SSDs at the core," said Axel Stoermann, Chief Technology Officer & VP at KIOXIA Europe. "By utilising SSD-based ANNS, we are reducing the reliance on costly DRAM while matching the performance needs of leading in-memory solutions – enhancing the performance range of large-scale RAG applications significantly."
