Sector News

Intel introduces GPU architecture for HPC/AI

Tue, Nov 19, 2019

At this week’s Intel HPC Developer Conference in Denver, Colrado, USA this week, Raja Koduri,  senior vice president, chief architect, and general manager of architecture, graphics and software at Intel (pictured), introduced a new category of discrete general-purpose GPUs optimised for artificial intelligence (AI) and high performance computing (HPC) convergence.

The Ponte Vecchio general purpose GPU is based on the Xe architecture.

“HPC and AI workloads demand diverse architectures, ranging from CPUs, general-purpose GPUs and FPGAs, to more specialised deep-learning neural network processors (NNPs), which Intel demonstrated earlier this month,” said Koduri.

Ponte Vecchio is architected for HPC modeling and simulation workloads and AI training. It will be manufactured on Intel’s 7nm technology and will be Intel’s first Xe-based GPU optimised for HPC and AI workloads. Ponte Vecchio will leverage Intel’s Foveros 3D and EMIB packaging and feature high-bandwidth memory, Compute Express Link interconnect and other IP in one package.

Intel also launched the oneAPI initiative to define programming for an AI-infused, multi-architecture world. oneAPI delivers a unified and open programming experience to developers on the architecture of their choice without compromising performance and eliminating the complexity of separate code bases, multiple-programming languages, and different tools and workflows. It preserves existing software investments with support for existing languages while delivering flexibility for developers to create versatile applications.

The oneAPI specification includes a direct programming language, powerful application programming interfaces (APIs) and a low-level hardware interface. Intel’s oneAPI beta software provides developers a comprehensive portfolio of developer tools that include compilers, libraries and analysers, packaged into domain-focused toolkits. The initial oneAPI beta release targets Intel Xeon Scalable processors, Intel Core processors with integrated graphics and Intel FPGAs, with additional hardware support to follow in future releases.

Developers can download the oneAPI tools, test drive them in the Intel oneAPI DevCloud.

http://www.intel.com

> Read More

Aetina quadruples AI power efficiency

Mon, Nov 18, 2019

To support the demand for more smart applications for AI embedded system and application developers, Aetina offers a series of system on modules (SoMs) built with Nvidia Jetson modules. The company has compared its Nvidia Jetson AGX Xavier edge computing platform with the RTX 2080 graphics card. Using the Aetina Jetson AGX Xavier AI computing platform, which has 512 CUDA computing cores and 11 TFLOPS of AI computing inference, the company compared the Nvidia RTX 2080 high-performance graphics card which provides 20.14TFLOPS computing performance, but it needs 225W for operation. However, with the Xavier platform, it only cost 30W for operation, bringing 0.367 TFLOPS/W, for a four-fold improvement in AI performance effciency, reports Aetina, compared to RTX 2080’s 0.09 TFLOPS/W. The Xavier platform supports various I/O extension, external high-speed Net card, and flexible match with different camera modules. As an SoM platform, the Aetina Xavier platform is easy to deploy at any edge position in the embedded IoT environment, lower the AI latency, and bring a perfect AIoT (AI of things) chain at the edge.

Using the software, the Aetina R&D team optimised the Openpose algorithm, to detect human skeletons and capture several people in real-time. With the data from the skeleton, it can define the pose with action, execute a higher-level application.

With Aetina Jetson AGX Xavier edge AI computing platform, high-performance with low power consumption can realise an AI application of collapse detection in the intelligent medical sector and also conduct customer behaviour analysis in smart retail applications.

Aetina was founded in Taiwan in 2012, as a support provider of high-performance GPGPU and Jetson edge AI computing solutions for embedded applications. The company focuses on the industrial market, providing industrial components and longevity of service.

http://www.aetina.com

> Read More

Rohde & Schwarz and LG verify wireless emergency alerts

Mon, Nov 18, 2019

New FCC enhanced targeting rules, due to take effect this month, will mean that providing the ability to geo-target residents who receive wireless emergency alerts (WEA) will soon be a mandatory requirement.

Rohde & Schwarz, in collaboration with LG Electronics Mobile Comm USA have verified WEA 3.0 features related to geo-fencing using an LG G8 ThinQ mobile smartphone. The enhanced targeting rules imposed by the Federal Communications Commission (FCC), require wireless providers to reach 100 per cent of the targeted geographic area specified by government issuers, with no more than a 0.1-mile (160m) overshoot.

WEAs are a critical part of the nation’s emergency readiness system for warning the public about dangerous weather, missing children and other emergencies. Geographic targeting is a necessary mobile device feature as alerting citizens not affected by a specific emergency could cause unnecessary panic and safety risks.

The Rohde & Schwarz solution consists of the R&S CMW500 radio communication tester and the R&S SMBV100B vector signal generator. The solution verifies mobile device signalling according to WEA 3.0 geo-fencing specifications. It can also cover 3GPP conformance and network operator test plans for LTE, WLAN, WCDMA and GSM location-based services.

Bryan Helmick, product manager for mobile radio testers at Rohde & Schwarz USA, explained, “Test systems that can verify device compliance are critical to the successful roll-out of the FCC’s enhanced targeting requirements. This effort between Rohde & Schwarz and LG Electronics helps to further the commercial rollout of WEA 3.0 and geo-fencing.”

http://www.rohde-schwarz.com

> Read More

Multi-band wireless communication LSI suits smart meters around world

Mon, Nov 18, 2019

A multi-band (sub-1GHz/2.4GHz) wireless communication large scale integration LSI chip, the ML7421, is optimised for applications requiring low power consumption over relatively long distances. Lapis Semiconductor, part of the Rohm group, says typical end users are smart meters, gas/fire alarms, smart agriculture and home/building security systems.

The use of smart meters has increased dramatically in Japan since 2015 and the Wireless M-bus system was launched in Europe. At the same time, there is a trend to collect and manage sensor data using wireless networks not only to optimise energy consumption and lighting in buildings but also for crime surveillance and disaster prevention. These wireless sensor networks are also increasingly used to improve productivity in agriculture.

The ML7421 covers the sub-1GHz frequency band (400MHz to 960MHz) and the 2.4GHz frequency band, providing universal compatibility. Previously, it was necessary to select wireless LSIs for each country and develop devices for them. By supporting the 2.4GHz band available world-wide, a single device can be deployed globally. Additionally, in environments where communication at 2.4GHz is unstable, long-distance Sub-1GHz communication is available. For this reason, 2.4GHz and Sub-1GHz can be used as bridged communication depending on the application or the environment. The ML7421 can be used in radio stations compliant to ETSI EN 300 200, FCC PART15, and ARIB STD-T66, T67, T108 with several packet handling functions based on Wireless M-Bus and IEEE802.15.4g.

It has stable wireless characteristics even under changing environmental parameters such as voltage and temperature fluctuations, says the company. Fluctuations over temperature (-40 to +85 degrees C) are only 0.5dB in Tx output power and 1.0dB in Rx sensitivity. A DC/DC converter, high-efficiency class-E power amplifier and a high-speed radio wave check function allow the average current consumption to decrease by 15 per cent compared with conventional Lapis products, to lower system power consumption and prolong battery life.

Improving delta-sigma ADCs makes it possible to achieve flexible data rate demodulation up to 300kbits per second for global usage, while improving receiver sensitivity. As a result, the ML7421 is suitable for equipment used outdoors such as smart meters and various IoT sensors. These stable characteristics make it possible to extend long-range communication further with high power amplifier, adds the company.

https://www.lapis-semi.com

> Read More

About Smart Cities

This news story is brought to you by smartcitieselectronics.com, the specialist site dedicated to delivering information about what’s new in the Smart City Electronics industry, with daily news updates, new products and industry news. To stay up-to-date, register to receive our weekly newsletters and keep yourself informed on the latest technology news and new products from around the globe. Simply click this link to register here: Smart Cities Registration