Proving the Business Case for the Internet of Things

Arm TechCon round up: News from San Jose

Steve Rogerson
October 16, 2018



Intel is teaming up with arch-rival Arm for securely onboarding both Intel and Arm IoT devices to any application or cloud framework, the two companies announced at this week’s Arm TechCon in San Jose, California.
 
The traditional manual onboarding process for IoT devices has multiple challenges. It typically takes more than 20 minutes per device and involves coordination among installation technicians, IT network and security operations and operational technology teams. The device identity and network access credentials are either painstakingly preloaded into the device at manufacturing or configured in the field from a standard image using insecure human processes.
 
Compounding the security issues is the proliferation of cloud-specific provisioning methods without a consistent hardware-protected device identity model. For IoT to scale to a trillion devices in less than two decades, this process must be faster, safer and more flexible.
 
Last October, Intel Secure Device Onboard was launched to enable a late binding approach to provisioning, where users could dynamically discover their target cloud platform for provisioning seconds after the device is powered on in the field.
 
The collaboration with Arm aims to extend this capability from Intel devices to include the Arm devices that commonly are deployed together by users. This strategic collaboration of two major ecosystems is designed to provide the industry with a more flexible provisioning method that can be natively enabled in devices.
 

 
“As a result, customers should be able to choose their onboarding systems of record without being locked into a single cloud provider’s provisioning method or a single device architecture,” said Lorie Wigle (above), vice president of Intel’s software and security group. “Flexibility can be built in before the device is purchased to onboard into any cloud ecosystem.”
 
Device management systems such as Pelion, cloud/on-premise IoT platforms and connected partner ecosystems can all benefit from increased variety of devices, lower cost and faster deployment. Device suppliers can manufacture to a single SKU that can be provisioned with user-specific credentials in the field rather than in the factory, reducing cost while decreasing time to market.
 
“Intel and Arm are simplifying one of IoT’s most complex and challenging barriers with regard to streamlining the manufacturing and security deployment workflows for IoT,” said Michela Menting, director of ABI Research. “This is an RoI win for the customer, who will be able to deploy both Intel- and Arm-based devices at a lower cost and with less friction between IT and OT, while at the same time retaining flexibility over their data and cloud partner choice until the deployment phase.”
 
Intel and Arm are seeking customer and ecosystem feedback on the prototypes and expect to engage pilot customers later this year.
 
“Intel’s collaboration with Arm allows us to progress a joint vision of any device, any cloud to span multiple device architectures,” said Wigle. “As we enter this accelerated growth phase for IoT, we will continue to collaborate with technology vendors to provide customers the protections they need.”
 
Other news from Arm TechCon
Arm announced strategic partnerships for its Pelion IoT platform with Intel, MyDevices and Arduino to increase IoT flexibility, simplicity and scalability for organisations.
 
Arm is building ecosystems by fostering a diverse team of partners to enable IoT to scale securely. In addition, Arm is announcing Mbed Linux OS, which builds on its Mbed OS IoT platform OS with more than 350,000 developers by enabling secure and rapid development and device management of IoT devices based on Cortex-A.
 
Arm also announced Neoverse, a dedicated roadmap and brand of purpose-built IP for 5G networks and next-generation cloud to edge infrastructure. Neoverse is designed for higher-levels of performance, security and scalability to enable best-in-class compute, networking and storage.
 
Lynx Software Technologies demonstrated LynxSecure, its separation hypervisor. LynxSecure simplifies safety certification processes for Arm-based designs, providing military-grade security coupled with virtualisation to multi-core Arm SoC-based designs. LynxSecure is available on the Xilinx Zynq Ultrascale+ MPSoC and the NXP S32V.
 
“LynxSecure supports mixed criticality development, from bare metal code running alongside full stack Linux, with secured communications capabilities between safely separated applications, allowing the mixing and matching of runtime environments on multi-core Arm-based SoCs,” said Lee Cresswell, vice president of global sales at Lynx Software Technologies.
 
LynxSecure simultaneously supports functional safety certification through hardware supported system partitioning and highly secure by design system resource management for real-time multi-core applications. Developers can run safety-critical OSes and applications next to general purpose OSes on a single SoC without compromising the performance, safety or features of either, because LynxSecure 6.0 provides secure separation and safe partitioning between these different environments and the devices that they use.
 
Software tools company IAR Systems and Secure Thingz, a domain expert in device security, embedded systems and lifecycle management, presented a shared vision of building a secure and sustainable future for connected devices.
 
“IoT security needs to be straightforward, scalable and sustainable,” said Haydn Povey, CEO of Secure Thingz. “Secure Thingz and IAR Systems have outlined a vision with a number of fundamental beliefs for helping our customers to secure intellectual assets, accelerate trustworthy product delivery and add value and services.”
 
Stefan Skarin, CEO of IAR Systems, added: “IAR Systems and Secure Thingz are working together to make superior security available for all by leveraging our respective heritages and technological leadership.”
 
ST Microelectronics is bringing cyber-protection to power-conscious connected devices with the STM32L5 microcontroller (MCU) series featuring the Arm Cortex-M33 core.
 
Building on the Cortex-M33, which boosts protection for small devices by integrating Arm’s TrustZone hardware-based security, the MCUs add further enhancements including flexible software isolation, secure boot, key storage and hardware cryptographic accelerators.
 
The MCUs also provide functionality, performance and long run-times powered by coin cells or energy harvesting. Consuming as little as 33nA in shutdown mode and achieving 402 ULPMark-CP in the EEMBC ULPBench, the series builds on the company’s expertise in low-power techniques such as adaptive voltage scaling, real-time acceleration, power gating and multiple reduced-power operating modes.
 
“The STM32L5 series with TrustZone and additional custom protection features considerably strengthens and hardens cyber-protection for small IoT devices,” said Ricardo De Sa Earp, general manager of STM’s microcontroller division. “Also featuring our unique energy-saving technologies, rich connectivity, and smart digital and analogue peripherals, these devices will be the first choice to host cutting-edge connected applications.”
 
Express Logic, provider of the X-Ware IoT platform powered by the Thread X RTOS, has integrated the ST Microelectronics X-Cube-STL software self-test libraries with the X-Ware IoT platform for developers using STM32 microcontrollers based on Arm cores.
 
X-Ware leverages the X-Cube-STL to help users more quickly navigate through the IEC 61508 Sil 2 and 3 safety integrity level certification process thus speeding delivery of STM32-based safety-critical products targeting markets such as industrial, motor control, factory automation, power generation and conversion, and medical.
 
Flex Logix Technologies demonstrated modular embedded FPGA (eFPGA) capabilities so SoC designers can use FPGA acceleration code with little effort.
 
"Modular FPGA enables the broader SoC software engineering community to add, change or customise hardware accelerators with a simple function call in their C/C++ software," said Geoff Tate, CEO and co-founder of Flex Logix. "This will allow designers to speed development of core designs, enabling acceleration to be added or modified as needed anytime in the process."