Object storage provider Scality has rolled out ADI (Autonomous Data Infrastructure), an enterprise-grade data management platform. Powered by policy-driven AI agents, it categorises data into four performance-, cost- and protection-based storage tiers.
Modern on-premises object storage now bears diverse AI workloads that exceed manual management capability. Use cases including training, inference, multimodal workflows, RAG and KV caching require distinct throughput, latency and governance. Storage must balance power limits, cyber resilience, data sovereignty and scalability without excessive cost and complexity.
Scality CEO Jérôme Lecat said: “The AI era hasn’t just changed how enterprises use data, it has exposed how badly the old storage model was broken. Scality ADI is a new operating model that autonomously tailors performance, protection and economics for every workload across the data lifecycle. It keeps GPUs productive, meets compliance requirements and sustains exabyte-scale sovereign control. We are not replacing what works; we are building what comes next.”
A typical Scality customer may run RING/RING XP on-premises storage and ARTESCA immutable backup, with multi-cloud support for AWS, Azure and GCP. They need sovereign data isolation, low-latency GPU access for cache-limited environments, and cost-effective tape archiving for cold data—all with guaranteed cyber resilience.
Legacy storage software struggles with such diverse demands, forcing fragmented deployments or inefficient single-tier flash storage. ADI acts as an AI-powered management layer for its multi-tier, single-namespace storage, optimizing performance, resilience, sovereignty and power efficiency for large-scale infrastructure.
It dynamically matches storage media, protection and performance to individual workloads, enabling AI computing, data protection and scalable growth without trade-offs.
ADI features four policy-based storage tiers:
-
Extreme performance tier: GPU-Direct TLC flash with S3 over RDMA and <50μs latency;
-
Hot tier: QLC and next-gen near-line SSDs with multi-TB/s RDMA bandwidth;
-
Warm tier: Next-gen NL-SSDs and NL HDDs via RDMA;
-
Cold tier: Tape and public cloud archival storage.
Scality recommends an optimized data distribution: 5% on Tier 1/2, 30% on warm tier and 65% on cold tier. Real-time power telemetry tracks energy consumption across systems, nodes and workloads to align performance with data center constraints.
Its built-in Guardian module delivers AI-driven operational intelligence, covering predictive maintenance, health monitoring, power control and threat detection. AI agents automate expansion, rebalancing, upgrades and lifecycle workflows.
Users can integrate custom AI tools via MCP to bypass native platform limitations. The solution adopts human-auditable logic: AI generates insights, users approve actions, and the system executes compliant workflows to cut manual workloads. ADI remains semi-autonomous.
Scality’s CORE5 suite ensures end-to-end data immutability, recoverability and auditability. Now available via global partners, ADI offers transparent source code, governed contributions and outcome-based SLAs for availability, performance, protection and power efficiency.
Comment
ADI currently lacks Nvidia STX KV Cache support, yet future integration is expected.
Beijing Qianxing Jietong Technology Co., Ltd.
Sandy Yang/Global Strategy Director
WhatsApp / WeChat: +86 13426366826
Email: yangyd@qianxingdata.com
Website: www.qianxingdata.com/www.storagesserver.com
Business Focus:
ICT Product Distribution/System Integration & Services/Infrastructure Solutions
With 20+ years of IT distribution experience, we partner with leading global brands to deliver reliable products and professional services.
“Using Technology to Build an Intelligent World”Your Trusted ICT Product Service Provider!
Votre message doit contenir entre 20 et 3 000 caractères!
French
