A European funded project has released an open-source framework which seamlessly simulates, in an integrated way, both the networking and the processing parts of cyber-physical systems (CPS), as well as cloud and high-performance computing systems.
Cyber-physical systems are supersets of embedded systems, integrating sensing, computation, control, and networking into physical objects and infrastructure. While IoT refers mainly to uniquely identifiable internet-connected devices and embedded systems, CPS refers to the combination of the multiple hardware devices and software (including computational) aspects of a system, together with its relationship with the physical world.
The European project, a three-year program which began in February 2015 and received 2.88 million euros (about $3.5 million) in funding from the European Commission under the Horizon 2020 program, addresses the lack of simulation tools and models for full system design and analysis. This is mainly because most existing simulation tools for complex CPS only efficiently handle parts of a system while mainly focusing on performance.
They also require extreme amounts of processing resources and computation time to accurately simulate the CPS nodes’ processing. Faster approaches are available, however as they function at high levels of abstraction, they cannot provide the accuracy required to model the exact behavior of the system under design to guarantee that it meets the requirements in terms of performance and/or energy consumption.
The new open source COSSIM framework, as it is called, developed in a R&D cooperation between STMicroelectronics with Politecnico di Milano, Telecommunication Systems Institute and Synelixis, was successfully evaluated on an advanced client-server visual search use case that detects instances of objects. The evaluation included co-simulation with an experimental embedded system made by ST Microelectronics.
COSSIM is a brand new open-source, highly parallel software framework that integrates a processing ‘full-system simulator’ with an innovative network simulator. Additionally, advanced power consumption and security measurement models have been developed and incorporated to the end framework. This results in a simulator that is orders-of-magnitude faster and more accurate, while reporting more CPS aspects than any existing solution. The COSSIM framework also incorporates a novel and efficient hardware acceleration process through the use of ST microcontrollers.
"The internet-of-things (IoT) market, in particular, urgently needs artificial intelligence (AI) functionalities to effectively capture its immense growth potential. A powerful CPS and cloud simulator can serve that need by supporting system designers in simulating and co-simulating much more effectively than by using traditional methodologies,” said Danilo Pau, Senior Member, Institute of Electrical and Electronics Engineers, and Senior Principal Engineer, Senior Member of the Technical Staff, STMicroelectronics. "The technology developed as part of the COSSIM project has the potential to boost system-design productivity and as an open-source framework will benefit from the contributions of the GitHub community."
Other project partners for this program were Search-Lab, Tecnalia, Chalmers University of Technology and Maxeler Technologies.
在线留言询价
型号 | 品牌 | 询价 |
---|---|---|
BD71847AMWV-E2 | ROHM Semiconductor | |
TL431ACLPR | Texas Instruments | |
MC33074DR2G | onsemi | |
CDZVT2R20B | ROHM Semiconductor | |
RB751G-40T2R | ROHM Semiconductor |
型号 | 品牌 | 抢购 |
---|---|---|
BP3621 | ROHM Semiconductor | |
IPZ40N04S5L4R8ATMA1 | Infineon Technologies | |
BU33JA2MNVX-CTL | ROHM Semiconductor | |
ESR03EZPJ151 | ROHM Semiconductor | |
TPS63050YFFR | Texas Instruments | |
STM32F429IGT6 | STMicroelectronics |
AMEYA360公众号二维码
识别二维码,即可关注