ضبط داده ها به طور مداوم بر روی سیستم های سریع زمان واقعی

The PCU-Project [1] launched for the enhancement of the vertical stabilisation system at JET required the design of a new real-time control system with the challenging specifications of View the MathML source2Gops and a cycle time of 50 View the MathML sourceμs. The RTAI based architecture running on an x86 multi-core processor technology demonstrated to be the best platform for meeting the high requirements. Moreover, on this architecture thanks to the smart allocation of the interrupts it was possible to demonstrate simultaneous data streaming at 5050 MBs on Ethernet while handling a real-time 100 kHz interrupt source with a maximum jitter of just 3 View the MathML sourceμs. Because of the memory limitation imposed by 32 bit version Linux running in kernel mode, the RTAI-based new controller allows a maximum practical data storage of 800 MB per pulse. While this amount of data can be accepted for JET normal operation it posed some limitations in the debugging and commissioning of the system. In order to increase the capability of the data acquisition of the system we have designed a mechanism that allows continuous full bandwidth (5656 MB/s) data streaming from the real-time task (running in kernel mode) to either a data collector (running in user mode) or an external data acquisition server. The exploited architecture involves a peer to peer mechanisms where the sender running in RTAI kernel mode broadcasts large chunks of data using UDP packets, implemented using the ‘fcomm’ RTAI extension [2], to a receiver that will store the data. The paper will present the results of the initial RTAI operating system tests, the design of the streaming architecture and the first experimental results.

The new vertical stabilisation system recently developed at JET has required the design of a real-time control system able to run at a cycle time of 50 View the MathML sourceμs with a very low jitter (¡2 View the MathML sourceμs). This challenge drove the development of a new high performance framework where to run real-time applications. The main objectives of the architecture were not only oriented towards a fast system but also to develop a flexible, easy to test and debug and exportable environment. High performance leads to the requirement of managing a large amount of data generated by the real-time application. Because of the limitation of the local memory available an advance and efficient mechanism of data streaming had to be designed. As a major requirement, the mechanism must provide high transfer rate without interfere with the hard real-time tasks. In the following section the real-time framework will be presented. The basic concepts of the implementation and the peculiarity of the system will be introduced. In Section 3 the streaming mechanism is described followed by the test results. In the last section the conclusion and the additional possible use of the framework will be treated.

The new real-time multiplatform framework developed at JET has demonstrated to be an optimal environment for developing applications at high performance. The modular approach adds flexibility and reusability together with an improved debugging and testing capabilities without compromising the real-time performance. The optimal results are reached with the multi-core processor technology where each hard real-time task and the less demanding supervisor (MARTe) are allocated on a different CPUs. The Linux-RTAI has shown to be the best platform thanks to a wise usage of the interrupts management. In order to cope with the huge flow of data coming from real-time applications, a new streaming mechanism has been developed for continuously transfer data from the hard real-time task to a collection target. An impressive 56 MB/s has been reached on a standard Gigabit network without compromising a 50 us cycle with jitter in the order of the hundreds of nanoseconds. The developed solution can be easily exported in an environment where the duration of the experiment doesn’t allow local data storing like ITER. Continuous sending of data during a long pulse can be solved with the new data recording system. Moreover with this architecture it could be possible to send a collection of data from the real-time tasks to offline post-processing systems. A chain of post elaboration can be attached to the real-time application providing elaborated data or visualisation without interfering with the experiment nor having to wait for the end of the experiment.