Arm to IoT devs: Go faster with our pre-made chip subsystems
In addition to cloud service, Arm also puts out its fastest Cortex-M CPU design yet
The belief that IoT would become Arm's main growth engine never played out as owner SoftBank Group thought it would, but the British chip designer is still doing what it can to keep IoT developers hooked with a bevy of new offerings meant to significantly speed up development.
On Tuesday, Arm announced its first major expansion of the Arm Total Solutions for IoT program, which consists of pre-integrated subsystem designs that take out the guesswork for chip designers, the Arm Virtual Hardware cloud service for testing Arm-based devices without needing physical silicon, and several software components developers can reuse across multiple devices.
Internet of Things: Arm crafts OS to rule them allARCHIVE
The expansion was accompanied by the reveal of the new Cortex-M85, which Arm said is its fastest Cortex-M CPU core blueprint yet for IoT devices powered by microcontroller units.
Mohamed Awad, vice president of IoT and embedded at Arm, told The Register that Arm devised the IoT program expansion based on feedback from developers and device makers, who are facing increasing pressure to bang out new and more powerful products faster.
"They're faced with this ever-increasing demand for higher performance, shorter time to market, increased security, simpler development, and then more secure processing. And that's the only way that they're going to continue to scale," he said.
We'd also like to think Arm is doing what it can to prevent IoT developers from moving to RISC-V, an open-source instruction set architecture that gained a higher profile after Nvidia's now-failed plan to acquire Arm was announced in 2020. Arm is also getting ready for a potential initial public offering, which could reportedly happen this year.
New subsystems for voice recognition, cloud-native edge devices
A key part of the Arm Total Solutions for IoT program is Corstone, a collection of pre-integrated subsystem designs that bring together Arm's CPU core blueprints with other important building blocks the company thinks is best for different kinds of applications.
These other kinds of building blocks can include component designs for power control, system control, system peripherals, interconnects as well as security-minded things like secure debug, secure enclave and the company's TrustZone Protection Controller.
The thinking is that since Arm knows which components work well together for various applications, why not just do that integration ahead of time so that chip designers at other companies don't have to waste precious time figuring it out.
Awad promised that developers can still make their own tweaks, and so far, this way of designing chips has helped companies tape out more than 200 designs since Corstone began three years ago.
"It's important to say that we work really hard to strike this crucial balance between providing all the essential ingredients for a complete solution but maintaining enough freedom and flexibility to allow the developers to get creative and add differentiation," he said.
Last October, Arm launched the Corstone-300 design for keyword recognition applications, and the company extended it today with two new designs for devices with higher performance needs: Corstone-310 for voice recognition applications and Corstone-1000 for cloud-native edge devices.
With the focus on voice recognition, Corstone-310 is well-suited for MCU-powered devices ranging from smart speakers and thermostats to drones and factory robots.
The Corstone-310 design makes use of Arm's new MCU-class Cortex-M85 core, for which the company is boasting a significant boost in both scalar performance and machine learning performance over the Cortex-M55. That performance gap is even wider when comparing it to the Cortex-M7.
"I like to call it the king of the M-class product line. It's bringing Cortex-M closer to the performance we're used to with Cortex-A but with the convenience, determinism and low power of an MCU," he said.
Cortex-M85 relies on Arm's Helium technology to boost machine learning and digital signal processing workloads. It also has enhanced security via Arm TrustZone technology, and it comes with a new architectural feature called pointer authentication and branch target identification to help developers gain level-two PSA certification for their chips.
- SoftBank aims to keep control of Arm after IPO – report
- Apple dev logs suggest 'nine new M2-powered Macs'
- Enemybot botnet uses Gafgyt source code with a sprinkling of Mirai
- 'Bigger is better' is back for hardware – without any obvious benefits
In addition to Cortex-M85, Corstone-310 also supports Arm's Ethos-U55 NPU blueprint as an option, which helps make it the company's highest-performance MCU-based subsystem design.
The Corstone-1000 subsystem design, on the other hand, is made for devices that require a higher level of performance and need to run on operating systems like Linux. This includes point-of-sale systems, edge gateways and high-end smart cameras.
As such, Corstone-1000 relies on both a Cortex-M core and a higher-performing Cortex-A core, which can range from the A32 to the A53. The subsystem design also comes with a secure enclave, which can be PSA-certified for processing sensitive information.
Corstone-1000 is part of Arm's SystemReady certification program, which ensures that it meets a set of hardware and firmware standards so that chips using the subsystem design "just work." It also supports Arm's Project Cassini, which aims to simplify cloud-native software experiences for devices running on Cortex-A-based processors.
Arm Virtual Hardware expands to more hardware
Another important way Arm is trying to make IoT development faster is with Arm Virtual Hardware, a service available on Amazon Web Services that allows people to develop software for processors without needing to have physical silicon in hand.
The news here is that the cloud service has now expanded to support the two new Corstone subsystem designs as well as seven Cortex-M cores, ranging from the M0 to the M33. This will make it much easier for developers to test out new software for devices running with Cortex-M designs, according to Award.
"By adding support for seven of our most popular Cortex MCUs, Arm Virtual Hardware is going to immediately support 80 billion devices that our ecosystem has already deployed. Software developers can immediately begin to develop test and verify their software across all those devices, without having to deal with all that hardware," he said.
The service also now supports Arm-based hardware from partner companies such as NXP Semiconductors, ST Microelectronics and the makers of Raspberry Pi. This expands the purview of IoT software testing for developers to even more devices, Awad added.
"Because they can leverage the scale of the cloud and don't need to build hardware farms, they can take advantage of all the benefits of modern development flows, things like continuous integration and streamlined ML DevOps and simplified security," he said.
The other new development is that Arm Virtual Hardware can integrate with existing development tools and environments. This includes Arm's own Keil MCU development kit as well as code repository GitHub and ML DevOps tools like Edge Impulse.
The last big update for the IoT program is that Arm is expanding its efforts to create a "consistent set of standards" across a variety of devices, so that developers can easily re-use various software components for standard things that don't need to be rewritten every time.
Arm is doing this with the first release of its Open IoT SDK Framework, which includes the community-driven Open-CMSIS Pack that comes with a set of standardized interfaces and interoperability technologies for Cortex-M-based devices. It also includes software for IoT applications like voice recognition and keyword recognition.
"By defining how they're accessed, we're enabling a level of software reuse and leverage for developers that just didn't exist before," Awad said. ®