NVIDIA Expands AI Storage Access with cuObject and SCADA Server SDK

NVIDIA Expands AI Storage Access with cuObject and SCADA Server SDK NVIDIA has announced the general availability of its cuObject client and server libraries and introduced the Scaled Accelerated Data Access (SCADA) Server SDK to standardize high-speed, RDMA-accelerated file and object storage for AI workloads. cuObject client and server libraries are now generally available, enabling zero-copy RDMA data transfers for object storage without passing through the host CPU. The xio-sig initiative is expanding to include cuObject alongside cuFile, supported by partnerships with Google Cloud and Microsoft. The new SCADA Server SDK, part of the Storage-Next initiative, allows storage providers to build servers responding directly to fine-grained, GPU-initiated I/O requests over RDMA. IBM demonstrated interoperability using a prototype SCADA server built on the SDK and integrated with IBM Storage Scale. cuObject Integration and xio-sig Expansion NVIDIA is expanding the xio-sig project to encompass cuObject in partnership with Google Cloud and Microsoft, addressing the lack of a common RDMA wire protocol for object storage. With the general availability of cuObject client and server libraries, AI accelerators can access object storage directly through zero-copy RDMA without traversing the CPU-controlled memory, reducing latency and CPU utilization. Repositories for cuFile and cuObject are established, with headers, protocol details, and implementation libraries slated for release following conformance testing. SCADA Server SDK and Storage-Next Through the NVIDIA Storage-Next initiative—a collaboration of more than 40 vendors and hyperscalers—NVIDIA is tackling small, fine-grained, GPU-initiated I/O demands. The new SCADA Server SDK allows storage vendors to build servers that handle requests from GPU-based SCADA clients and return data over RDMA using local or remote storage. Alongside tools like the Storage Lender Service and a SCADA deployment CLI, IBM has shown the ecosystem's viability via a prototype integrating the SDK with IBM Storage Scale.

NVIDIA推出cuObject与SCADA Server SDK以扩展AI存储访问能力 NVIDIA宣布扩大xio-sig倡议并将NVIDIA cuObject与cuFile并列，同时正式推出cuObject客户端与服务器端库及SCADA Server SDK，旨在通过RDMA零拷贝加速和开放标准，提升GPU对文件和对象存储的访问效率与互操作性。 NVIDIA与Google Cloud、Microsoft合作将cuObject纳入xio-sig，cuObject客户端和服务器端库已全面上市（GA）。 cuObject支持通过RDMA实现绕过CPU的零拷贝数据访问，降低延迟并提高吞吐量，治理文件与相关代码正在筹备开放。 新推出的SCADA Server SDK支持存储提供商构建响应GPU端请求的服务器，IBM已基于Storage Scale展示了互操作原型。 在涵盖40多家厂商的NVIDIA Storage-Next计划下，SCADA为加速器发起的高并发、细粒度I/O提供了软件基础设施基础。 cuObject加入xio-sig与RDMA互操作性 针对AI负载中缺乏统一RDMA对象存储传输协议的痛点，NVIDIA与Google Cloud和Microsoft合作，在xio-sig中引入了cuObject并宣布其库正式全面可用。cuObject允许AI加速器借助RDMA直接访问文件与对象存储，无需通过服务器CPU内存中转，从而降低延迟和CPU占用率。目前xio-sig仓库已建立，相关协议规范、头文件及代码将在通过一致性测试后发布，微软亦计划加入董事会以促进存储I/O互操作。 SCADA Server SDK与Storage-Next生态进展 针对加速器发起的细粒度、小规模I/O请求，NVIDIA通过联合超40家厂商的Storage-Next计划推动GPU驱动存储标准。作为该体系的软件基础设施，新发布的SCADA Server SDK搭配Storage Lender Service与命令行工具，使存储伙伴能够构建响应GPU客户端请求的服务器并通过RDMA交付结果。IBM Storage已基于该SDK展示了与IBM Storage Scale集成的SCADA原型系统，可用于语义搜索、推荐系统等海量数据场景。

AI infrastructure engineers, storage developers, and cloud service providers need fast and secure access to high-capacity file and object storage to support AI workloads. AI workloads increasingly require high-speed data access for training, fine-tuning, inference context, tool calls, searches, and database lookups. Much of this data lies in files and objects stored both on-premises and in the cloud. Compute accelerators—including GPUs, TPUs, and XPUs—need remote direct memory access (RDMA) that uses NIC- or DPU-accelerated data transfers (like NVIDIA ConnectX NIC or NVIDIA BlueField DPU) and doesn’t copy the data through the server’s CPU-controlled memory. The need for RDMA-accelerated, zero-copy data transfers grows with faster GPU architectures. Developers building direct access to file and object storage have had to navigate different APIs and protocols across providers. Object storage over RDMA has also lacked a common wire protocol, leaving developers to support provider-specific integrations or rely on traditional access methods. The expansion of xio-sig and the new Scaled Accelerated Data Access (SCADA) Server SDK offer two ways to build more interoperable storage access for AI workloads. ## New developments in xio-sig and SCADA NVIDIA is expanding xio-sig to include **NVIDIA cuObject** alongside cuFile, in partnership with Google Cloud and Microsoft. It is also announcing the general availability of cuObject client and server libraries. Developers can use cuObject’s APIs and RDMA wire protocol to build accelerated object-storage applications and servers, while xio-sig provides a path toward enabling the cuObject client to interoperate with any server-side implementation that adheres to the wire protocol. A new SCADA Server SDK enables storage providers to build SCADA servers that respond to GPU-initiated requests from SCADA clients. IBM has shown interoperability with a prototype that integrates SCADA and IBM Storage Scale. Together with the xio-sig expansion, these efforts give AI developers, storage providers, and cloud services more ways to build and use accelerated storage through shared APIs and protocols. There is also an emerging class of AI data access involving small, fine-grained I/O requests initiated by accelerators that don’t fit traditional storage media or protocols. Through the NVIDIA Storage-Next initiative, NVIDIA is leading a group of over 40 vendors and customers, including NAND vendors, controller vendors, storage providers, hyperscalers, and application developers, to define how GPU-driven storage should work and turn those advancements into interoperable, open industry standards. The software infrastructure supporting high-throughput, fine-grained, GPU-initiated storage access is SCADA. The [FMS blog on open-source cuFile APIs, Storage-Next, and SCADA](https://blogs.nvidia.com/blog/ai-storage-fms/) provides background on these efforts. ## cuObject support in xio-sig The general availability of cuObject libraries provides AI application developers, open source framework developers, storage providers, and storage consumers with standardized methods for accelerated AI data access using both file and object protocols, as shown in Figure 1. AI accelerators can access file and object storage using RDMA, without routing data through the server CPU, enabling higher throughput, lower latency, and reduced CPU utilization for data writes and reads. Building on its involvement as a maintainer for cuFile in *xio-sig*, Google Cloud is evaluating expanding its participation for cuObject, reflecting its focus on high-performance cloud file and object storage. Microsoft also looks forward to joining the *xio-sig* Board to improve interoperability in storage I/O. ## xio-sig progress The repository structure is set up for each of cuFile and cuObject. Headers for cuFile and cuObject, the cuObject wire protocol, and library implementation code for libxFile and xFilekernel code will be shared once the production-ready stack passes conformance tests. The governance documents are currently under review by pending Board Members. ## NVIDIA Storage-Next and SCADA expansion Within the Storage-Next initiative, the new SCADA Server SDK enables storage partners to build servers that receive requests from GPU-based SCADA clients, fulfill them using local or remote storage, and deliver the results over RDMA. The effort also includes the [Storage Lender Service](https://github.com/NVIDIA/storage-lender) and a SCADA command-line utility for configuring and deploying SCADA. These components support storage-provider servers that can interoperate with SCADA clients. IBM Storage has demonstrated a prototype in which a SCADA client sends requests to their initial version of the Storage Scale SCADA server, built on the new SCADA Server SDK. This shows how storage vendors can collaborate with NVIDIA to build an ecosystem for SCADA’s accelerated, GPU-initiated storage access, paving the way for software infrastructure that could support access to large datasets for applications such as semantic search, recommender systems, and fraud detection. ## Get started with cuObject and xio-sig With xio-sig expanding to include NVIDIA cuObject and cuObject libraries now generally available, storage partners, providers, and consumers can start using cuObject and prepare to contribute to the community’s work on interoperable APIs and protocols for both cuFile and cuObject. - Watch the [xio-sig](https://github.com/xio-sig) repository for updates - NVIDIA [FMS blog](https://blogs.nvidia.com/blog/ai-storage-fms/) about xio-sig, cuFile, and Storage-Next - NVIDIA [presentation on SCADA](https://youtu.be/OZloQQZGmQQ?si=g8LTmGPnfXIYOdCb) at the 2024 OCP Summit - Check out and try the OSS [Storage Lender Service](https://github.com/NVIDIA/storage-lender) - GPUDirect Storage (GDS) [documentation](https://docs.nvidia.com/gpudirect-storage/) - Documentation for [cuFile APIs](https://docs.nvidia.com/gpudirect-storage/api-reference-guide/index.html#) - Documentation for [cuObject](https://docs.nvidia.com/gpudirect-storage/cuobject/index.html) - Link to [cuObject Server 2.0.0](https://developer.nvidia.com/cuobjserver-downloads) - NVIDIA CUDA 13.4 [preview downloads](https://developer.nvidia.com/cuda-13-4-0-download-archive) *See the notice regarding software product information.*