Kioxia Corporation has announced an update to its KIOXIA AiSAQ™ software. This new open-source release introduces flexible controls. These controls allow system architects to define the balance between search performance and the number of vectors.
The vectors are opposing factors within the fixed capacity of SSD storage in the system. The update enables architects of RAG systems to fine-tune the optimal balance. They can do this for specific workloads and their requirements without any hardware modifications.
KIOXIA AiSAQ software was first introduced in January 2025. It utilizes a novel approximate nearest neighbor search (ANNS) algorithm optimized for SSDs. This eliminates the need to store index data in DRAM.
By enabling vector searches directly on SSDs and reducing host memory requirements, KIOXIA AiSAQ RAG technology allows vector databases to scale. They can do this without being constrained by limited DRAM capacity. When the installed SSD capacity in the system is fixed, increasing search performance requires more SSD capacity consumed per vector.
Flexible tuning in AI workflows
This results in fewer vectors. To maximize the number of vectors, SSD capacity consumption per vector needs to be reduced.
This results in lower performance. The optimal balance between these two opposing conditions varies depending on the specific workload. To find the right balance, KIOXIA AiSAQ software introduces flexible configuration options.
The latest update enables administrators to select the optimal balance for various workloads within the RAG system. The update also makes KIOXIA AiSAQ technology a suitable SSD-based ANNS for other vector-hungry applications such as offline semantic searches. With the growing demand for scalable AI services, SSDs offer a practical alternative to DRAM.
They can manage the high throughput and low latency that RAG systems require. KIOXIA AiSAQ software meets these demands efficiently. It allows for large-scale generative AI without being limited by memory constraints.
By releasing AiSAQ software as open-source, the company reinforces its commitment to the AI community. It also promotes SSD-centric architectures for scalable AI.
Deanna Ritchie is a managing editor at DevX. She has a degree in English Literature. She has written 2000+ articles on getting out of debt and mastering your finances. She has edited over 60,000 articles in her life. She has a passion for helping writers inspire others through their words. Deanna has also been an editor at Entrepreneur Magazine and ReadWrite.
