English
Scality has unveiled ADI (Autonomous Data Infrastructure), an intelligent data platform tailored to resolve pervasive structural challenges within modern enterprise AI pipelines. Serving as Scality’s next-generation core solution built on its mature object storage foundation, ADI extends the firm’s infrastructure stack with autonomous data management capabilities. It is engineered to deliver secured, governable data orchestration for AI training and inference workloads running on distributed enterprise clusters.
Scality Core Storage Foundation
As AI workloads expand from isolated model training to iterative data processing and automated analytical tasks, maintaining compliant, sovereign data control across operational cycles has grown extremely vital. Under conventional storage architectures, sensitive enterprise data faces fragmented management due to siloed storage pools and disjointed security policies. This forces repeated data migration and redundant copy generation, driving up latency, administrative overhead and operational costs. Scality defines this inefficient workflow as the "data fragmentation tax", a systemic flaw that amplifies sharply in exabyte-scale business environments.
ADI is engineered to mitigate this pain point via a unified, autonomous data management layer capable of exabyte-scale pooling with optimized access latency. By consolidating heterogeneous data assets throughout AI workflows, the platform cuts redundant data movement and elevates overall infrastructure utilization. Internal benchmark data from Scality verifies streamlined data ingestion latency under enterprise-grade concurrency. In a typical deployment equipped with multi-rack GPU clusters and long-context dataset processing, resource utilization jumped from approximately 52% to over 91%, translating to substantial annual infrastructure expenditure reductions.
Scality’s product executives stated that data fragmentation remains negligible in small-scale test environments yet turns into a prominent operational bottleneck at enterprise scale. As business AI clusters expand, repetitive data replication and cross-pool migration incur higher labor costs and compliance risks, making autonomous governance systems indispensable for secure and sustainable AI deployment.
Addressing the Performance-Sovereignty Tradeoff
Legacy enterprise infrastructure forces teams to compromise between processing performance and data sovereignty. High-speed storage tiers such as all-flash arrays deliver low-latency access but come with high deployment costs and limited compliance customization. In contrast, archival storage systems offer massive capacity and robust governance but suffer from slow data retrieval, making them incompatible with dynamic AI iteration and real-time data analytics tasks.
Enterprise Flash Storage
ADI bridges this industry gap by introducing an autonomous middleware tier that balances accelerated data access and rigorous sovereign governance. Natively compatible with modern GPU accelerators and integrated with Scality Guardian intelligent scheduling engine, the solution enables entire enterprise clusters to access unified data pools with AI-optimized transmission speeds. This design eliminates cumbersome data duplication between isolated storage partitions, lowering service latency and elevating cluster throughput.
Intelligent Storage Controller
Architecture Optimized for AI Workloads
Tailored exclusively for enterprise AI data pipelines, ADI fits into the intermediate governance layer of Scality’s modern data hierarchy framework. Built on disaggregated storage infrastructure, it achieves exabyte-level expansion while maintaining optimized I/O latency, successfully decoupling storage scalability from GPU compute resources.
The system abandons rigid traditional storage architectures, migrating data straight across heterogeneous storage media via end-to-end offload transmission. This cuts out performance overhead brought by redundant protocol conversion, metadata redundancy and manual intervention—common bottlenecks in legacy file and object storage systems.
Source: Scality Official
Key architectural optimizations include native deployment on ARM64 and x86 hybrid servers, embedded directly within the storage layer to reduce dependence on third-party middleware. All data scheduling and migration adopt autonomous intelligent orchestration, bypassing repetitive manual configuration workflows. Additionally, ADI utilizes customized block sizes ranging from 4 MB to 16 MB, which are optimized for AI batch processing characteristics instead of conventional fixed storage blocks. It supports cutting-edge high-speed interconnection fabrics such as advanced Ethernet and PCIe Gen5, facilitating near wire-speed data transmission across enterprise clusters.
Availability
Scality ADI is now commercially available for global enterprise production deployment.
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!
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!