Powerful, Efficient Computational Storage
The volume of data we generate is expected to grow by an astonishing 27% a year. Applications such as the IoT are generating a volume of data that is becoming harder to manage. To further accelerate IoT deployments, data insights must be generated as quickly as possible. Clearly, with these challenges, data processing can no longer only happen in the cloud.
Computational storage, also known as in-situ processing or in-storage compute, brings high-performance compute to traditional storage devices. Data can be processed and analyzed where it is generated and stored, empowering users and organizations to extract valuable, actionable insights at the device level.
Secure, Energy-Efficient Smart Storage
Advantages of a computational storage drives (CSD) over traditional storage systems include:
- Latency reduction: analyzing data where it is created and stored allows results and insights to be generated in near real-time.
- Bandwidth and power efficiency: reducing the workload on servers in the data center leads to lower operating and cooling costs.
- Security and privacy compliance: data remains on the drive for aggregation and processing, helping to mitigate communication vulnerabilities.
- Flexibility and scalability: easily scale performance with added storage infrastructure and produce flexible solutions for different storage applications.
Arm offers the easiest, fastest, and most cost-effective computational storage solution with our low-power, high-performance processors and a software stack and developer tools from the Linux ecosystem.
Get Started on a Computational Storage Drive
This easy-to-read guide provides an introduction to computational storage for architects, developers, and manufacturers. It discusses real-world examples of how computational storage is deployed today by Arm partners, including in NVMe-based drives, and what to consider when implementing it in storage devices.
Computational Storage Use Cases Across Industries
As the amount of data being generated continues to explode, meaningful insights must be derived as quickly as possible. These common use cases demonstrate the significant impacts of implementing computational storage.
Powerful compute on storage devices allows database searches and operations to be performed directly on the data, facilitating timely insights and data security.
The ability to perform compute and machine learning without having to move large image files reduces bandwidth and latency, enabling real-time identification and classification.
High-performance edge compute performed closer to data helps ensure the lowest time-to-value and improves insights, even for the most complex data workloads.
Arm Technology for Smart Storage Applications
Cortex-R82 supports Arm Neon technology for accelerating ML workloads and a memory management unit (MMU) for enabling rich operating systems (OS). This high-performance Cortex-R82 processor supports the evolution of data storage.
Arm Cortex-A55 is our most powerful and efficient mid-range Cortex-A series processor. With a flexible design, it is widely deployed in cost-effective, energy-efficient storage devices.
Arm CoreLink NIC-450 is deployed across a wide variety of storage applications. It offers a highly configurable, low-latency topology, designed for building high-performance and scalable SoC connectivity.
What Arm Partners Are Saying
“We believe that more than 4TB of data will be generated daily by just one autonomous vehicle. Pre-processing will be essential in order to decide which data should be uploaded... [and] computational storage can be a great tool to manage this pre-processing in the most effective and productive way.”
Talk with an Expert
Learn more about how Arm’s technology, support and ecosystem can help you successfully implement computational storage in your devices and systems.
Arm Technology Moves Compute Closer Than Ever to Data
Arm is enabling compute as close as possible to where data is generated, reducing latency to maximize data time-to-value.
Computational Storage Enables Real-Time Data Processing
Computational storage can help realize the potential of emerging technologies, including AI and ML, by enabling insight and value generation closer to the data.
Cortex-R82: Real-Time 64-bit Performance & Applications Processing
The Cortex-R82 processor provides high performance, more addressable space, and can run Linux, introducing new capabilities for next-gen storage devices.
- Highest performance Arm Cortex-R processor to power the future of computational storage
- ScaleFlux computational storage makes Percona MySQL faster
- Computational storage: What is it and what are its key use cases?
- Computational storage winds its way towards the mainstream
- NGD Systems delivers industry's first ASIC-based computational storage platform in a NVMe SSD