On This Page: [hide]
NVM Express (NVMe) has introduced the NVMe Computational Storage Feature, a pioneering development that establishes a standardized and vendor-neutral framework.
NVMe Computational Storage Feature: Revolutionizing Data Processing Efficiency
This framework seamlessly connects applications with NVMe Computational Storage devices, merging compute and storage services. The launch of this feature signifies a substantial leap forward in how enterprises and hyperscale data centers handle and process data.
Bill Martin, Co-Chair of the NVMe Computational Storage Task Group and a Board Member, emphasized the importance of this release, noting that the introduction of NVM Express Computational Storage marks a pivotal step in adapting to the dynamic demands of the storage industry. This standardized approach fosters an open, interoperable ecosystem, offering the potential for reduced ownership costs and enhanced performance through computational offloading.
At the heart of Computational Storage is its capability to minimize data movement by enabling computation directly on the storage device. This innovation proves particularly advantageous for latency-sensitive applications such as databases, AI suggestion models, and content delivery networks, where rapid response times are paramount.
Introducing Two New Command Sets
The NVM Express Computational Storage Feature introduces two groundbreaking command sets:
- Computational Programs Command Set. This provides a host-driven, modular framework for computational programs, enhancing data processing efficiency.
- Subsystem Local Memory Command Set. This enables access to memory within an NVM subsystem. It is facilitated through computational programs and NVMe transport, accessible via NVMe I/O commands.
When integrated, these features signify a substantial advancement in computational storage technology. They hold the potential to elevate data processing performance and efficiency across various sectors. This innovation is anticipated to profoundly impact the future development of data processing and storage, offering tangible benefits in terms of performance, cost-effectiveness, and scalability for enterprises and data centers alike.