Highlights:

  • Cortex-R82 is the first Arm processor that combines both MMU-based and real-time contexts in a single core.
  • Cortex-R82 is the first Armv8-R 64-bit processor that offers enhanced performance to run new workloads such as ML.
  • This Arm processor is accelerating the computational storage revolution with 40 address bits.

Arm recently launched its highest performance 64-bit Cortex-R processor for smart storage applications. The processor is with memory management unit (MMU), enabling real-time and rich operating systems (OS), such as Linux, on a similar cluster or core. A 64-bit Cortex-R processor – the first arm with Cortex-R82 can address about 1 TB of dynamic random-access memory (DRAM) for next-generation of storage devices. The chip also has an optional ‘NEON’ acceleration. It is ideal for built-in storage solutions, solid-state drives (SSDs), and hard-disk drives (HDDs), as well as computational storage applications.

This processor plays a vital role while running stored data task-specific apps to offload the host and get faster results. For use cases, Arm said, “include video transcoding, database acceleration, and real-time data analysis.”

“Computational Storage is emerging as a critical piece of the data storage puzzle because it puts processing power directly on the storage device, giving companies secure, quick, and easy access to vital information,” commented Neil Werdmuller, Arm’s Director of Storage Solutions during SNIA Storage Controller session.

Recently, the company stated, “About 85 percent of hard disk drive and SSD controllers already use ARM processors for size, ecosystem and power-efficiency.”

The R82 is Arm’s first 64-bit Cortex embedded real-time processor and is up to two times faster than the previous Cortex-R8 generation. The MMU combines the physical memory space with space on a storage drive to offer a larger-than-physical memory virtual space. An R82-based storage controller can run numerous applications. For instance, in silent periods, it might use idle cores to run data transcoding, data analytics, or machine learning (ML) code. Arm said, “The R82 is 14 times faster than its Cortex-R8 for neural network workloads per cycle.”

The Cortex-R82 processor allows novel capabilities of future storage devices, bringing:

  • Enhanced flexibility
  • Faster response time and reduced latency
  • Augmented capacity

Arm NEON technology can be used to accelerate the deployment of ML workloads and compute-intensive tasks. NEON offers Single Instruction Multiple Data (SIMD) capability to get more work done. Further, it helps to boost performance when similar operations need to be performed on multiple data objects as in graphics work and digital signal processing. As per Arm, NEON can be exclusively used for video and audio processing, deep learning, facial and voice recognition, and computer vision.

Werdmuller said, “The R82’s features will allow storage applications to run new workloads like machine learning at a lower latency.” He thinks internet-of-things (IoT) edge applications might use storage drives with this capability.

Next-Generation Data (NGD), a supplier of computation-on-storage, might welcome the Cortex-R82.

Arm global goals’ impact

Arm has recently collaborated with other key organizations to work toward the achievement of the United Nations Sustainable Development Goals for 2030. Arm is one of the leading technology providers of silicon IP for the intelligent System-on-Chips at the heart of billions of devices.