Nvidia extends LLM memory with tiered KV caching and Dynamo engine
Nvidia GPUs store vectors as key-value pairs in a large language model (LLM) memory cache – KV cache – which is tiered out in a multi-level structure ending with...
VAST Data cracks into HPC with Doudna supercomputer win
VAST Data is supplying an AI-focussed storage system for the Doudna supercomputer, sharing the storage billing with IBM’s Storage Scale and its traditional high-performance computing (HPC) parallel file system.
Doudna...
Kioxia tunes SSD-based vector search for RAG workloads
Kioxia has tweaked its AiSAQ SSD-based vector search by enabling admins to vary vector index capacity and search performance for different workloads.
RAG (retrieval-augmented generation) is an AI large language...
Storage news ticker – June 30
Abstract Security announced LakeVilla, a new product aimed at offering customers a cloud-native cold storage facility built for long-term security telemetry retention that delivers compliance-ready, highly accessible storage at...
VDURA takes aim at VAST, WEKA, DDN with AI-HPC storage reboot
VDURA CEO Ken Claffey believes that the company should be classed alongside DDN, VAST Data, and WEKA as an extreme high-performing and reliable data store for modern AI and...
Micron touts 2600 QLC SSD with TLC-class write performance
The new 2600 client QLC SSD from Micron dynamically optimizes its cache to get QLC flash writing like TLC.
NAND has four basic formats for the number of bits in...
NetApp names Zscaler CTO Syam Nair as new chief product officer
NetApp has recruited Zscaler CTO Syam Nair to be its new chief product officer, replacing the departing Harvinder Bhela.
Bhela joined NetApp in January 2022 as EVP and CPO and...
Spectra Logic to archive Rubin Observatory’s 500 PB cosmic time-lapse
The first images have come from the Vera C. Rubin telescope in Chile showing an unprecedented view of the Southern Hemisphere’s night sky and the start of a ten-year...
Kioxia builds go-faster single-port datacenter SSD
Kioxia has refreshed its CD8P product line, a single-port SSD using BiCS5 112-layer 3D NAND, with the speedier CD9P line and its denser BiCS8 218-layer NAND.
Capacities of these PCIe...
Hitachi Vantara iQ product recognized by GigaOm
Hitachi Vantara has been named a Leader and fast Mover in a GRadar for High-Performance Storage Optimized for AI Workloads report by the GigaOm research business.
The background here is...
Pure debuts Enterprise Data Cloud with Fusion updates and new FlashArray systems
Pure Storage unwrapped its previously trailed Enterprise Data Cloud architecture in Las Vegas this week as it kicked off its Accelerate 2025 customer event.
The platform's debut delivers on its...
WEKA builds NeuralMesh architecture for exascale AI workloads
WEKA has announced a NeuralMesh transformation of its parallel file system software to accelerate AI at scale.
The basis of this is that traditional storage architectures have not been designed...