A new implementation of data process in J-TEXT real-time framework
Autor: | Yichang Pan, M. Chen, M. Zhang, Wei Zheng, Guozhen Zheng, J-Text Team, Y Li |
---|---|
Rok vydání: | 2018 |
Předmět: |
Data processing
010308 nuclear & particles physics Computer science business.industry Integration testing Mechanical Engineering Distributed computing Multiple buffering Linux kernel Modular design 01 natural sciences 010305 fluids & plasmas Data sharing Nuclear Energy and Engineering Data logger 0103 physical sciences General Materials Science Semaphore business Civil and Structural Engineering |
Zdroj: | Fusion Engineering and Design. 129:78-82 |
ISSN: | 0920-3796 |
DOI: | 10.1016/j.fusengdes.2018.02.060 |
Popis: | J-TEXT real-time framework (JRTF) is the latest development in real-time technology for developing plasma control system and other real-time applications on tokamaks by J-TEXT. It is powered by Linux open source software, fully embraced the object-oriented programming technology, and the modular design makes it a promising candidate for developing the prototype of PCS of China Fusion Engineering Test Reactor (CFETR). However, the data sharing between different application blocks (AB) could be time-consuming that brings in un-deterministic performance. In this paper, we first introduce the framework and the test device J-TEXT. After that, new data sharing method based on the mutex and double buffering are designed. Note that, kernel module programming which runs directly from Linux kernel space is also introduced to try to find some deterministic performance solutions to real-time data sharing process. In addition, in order to handle data more systematisms with searching from all the logs emitted by many ABs efficiently, data logging based on databases technology and a produce-consume model is lead into. This is also a key to solve simultaneously problem which may have an influence on the real-time performance. Then, an integration testing has been conducted to verify the realistic performance after bringing such method of data process to guarantee that data process can satisfy nowadays real-time and non-real-time needs in fusion aspects. Several comparing experiments also reveal the advantages and disadvantages of such process. |
Databáze: | OpenAIRE |
Externí odkaz: |