Tagged articles
AI storage
13 articles · Page 1 of 1
The FMW 2026 AI Storage Chip Map structures the AI storage ecosystem into three layers — system, device, and media/chip — listing key vendors from cloud providers and SSD controllers to HBM and emerging memories, while highlighting trends such as Chinese vendor rise, AI-centric tech hotspots, and open interconnect standards.
AI storageCXLFMW 2026
0 likes · 10 min read

The 2026 Future of Memory and Storage conference reveals a decisive industry pivot from traditional SSD scaling to AI-centric memory architectures, highlighted by Samsung's zHBM stacking, Kioxia's GPU-direct XL-FLASH SSD, SK Hynix/WD's open HBF standard, and CXL-enabled memory pooling and tiering for next-generation AI data centers.
AI storageCXLFMS 2026
0 likes · 10 min read

The article examines how AI models expose storage bottlenecks, introduces High Bandwidth Flash (HBF) as a 3D NAND‑based, HBM‑like stacked memory that combines subarray parallelism and a dedicated logic die, evaluates its technical principles, industry positioning, benchmark results showing 2.69× performance‑per‑watt gains, supply‑chain landscape, emerging standards, and the challenges that must be solved before commercial deployment.
AI storageCBA architectureHBM
0 likes · 10 min read

The article analyzes the limitations of traditional NAS for AI agents—scale ceiling, isolation gaps, and security blind spots—and explains how AgenticFS redesigns the storage paradigm by giving each agent an independent, zero‑trust workspace that scales linearly across regions.
AI storageAgenticFScloud native storage
0 likes · 13 min read

Baidu's PFS L3 rearchitects parallel file storage for AI production loads with a hybrid data engine (Aries + BlockServer), scalable metadata foundation (MetaDB), high-performance client (RapidFC), RDMA network tuning, transparent tiering to object storage, and AgenticFS for multi-tenant Agent workloads, proven in RL training saving 30k GPU-hours weekly.
AI storageAgenticFSAries
0 likes · 29 min read

3FS, DeepSeek’s high‑performance parallel file system, is engineered for AI workloads, offering ultra‑low latency, high‑throughput storage via RDMA, CRAQ consistency, and seamless cloud‑native integration, with detailed architecture, deployment steps, performance benchmarks, and cost‑saving strategies for large‑scale model training and inference.
AI storageRDMAdistributed file system
0 likes · 28 min read

This article analyses the emergence of AI‑focused storage, detailing its ultra‑high bandwidth, concurrency, scale and low‑latency characteristics, the architectural shift from layered to fused designs, the specific performance and data‑management demands of training and inference, and a three‑phase roadmap for future storage innovations.
AI storageGPU accelerationHigh-Performance Computing
0 likes · 12 min read

This article analyzes the evolution of distributed storage—from traditional compute‑storage separation to edge‑centric, AI‑enabled architectures—covering service models, key technologies such as CXL and erasure coding, reliability strategies, performance optimizations, vendor landscapes, and emerging green and intelligent trends.
AI storageCXLNVMe-oF
0 likes · 13 min read


AntDataMar 7, 2025 · Artificial Intelligence
This article examines the challenges of AI storage for massive models, describes Ant Group's multi‑cloud, high‑availability PCache architecture, and details its GPU‑mixed deployment, metadata services, data‑link optimizations, and performance results that enable petabyte‑scale training with low cost and high stability.
AI storageMulti-cloudPCache
0 likes · 19 min read

The article provides a comprehensive English overview of 3FS, an AI‑focused distributed file system that leverages FoundationDB for metadata, CRAQ for chunk replication, and a hybrid Fuse/native client architecture, detailing its design, components, fault handling, and performance considerations for large‑scale training workloads.
AI storageCRAQ replicationFoundationDB
0 likes · 25 min read

At the CCF China Storage Conference 2024 in Guangzhou, Alibaba Cloud’s research team presented the ALS (ALink System) ecosystem, discussed Scale‑Up interconnect protocols, and demonstrated multi‑layer storage innovations for AI workloads, highlighting hardware‑software integration, high‑bandwidth networking, and future CXL/PIM research.
AI storageALSStorage Innovation
0 likes · 7 min read

Facing petabyte‑level data, billions of small files, and the need for POSIX, S3, and HDFS compatibility, Didi designed a new generation of non‑structured storage—OrangeFS—by analyzing internal systems, combining multiple storage solutions, reusing GIFT technology, and implementing a high‑performance metadata service, multi‑protocol fusion, and robust scalability features.
AI storageMulti-Protocolbig data
0 likes · 27 min read

This article describes Zhihu's hybrid cloud architecture—including offline, online, and GPU data centers—its self‑built UnionStore cache, the performance and latency challenges faced during large‑scale AI model training, and the subsequent evaluation and migration to Alluxio community and enterprise editions to achieve higher throughput, stability, and lower operational overhead.
AI storageAlluxioOperations
0 likes · 14 min read
