Immediate data access system for LHD experiments

Several kinds of computer systems are used to perform large helical device (LHD) experiments, and each produces its own data format. Therefore, it has been difficult to deal with these data simultaneously. In order to solve this problem, the Kaiseki server was developed; it has been facilitating the unified retrieval of LHD data. The data acquired or analyzed by various computer systems are converted into the unified ASCII format, or Kaiseki format, and transferred to the Kaiseki server. With this method, the researchers can visualize and analyze the data produced by various kinds of computers in the same way. Because validations are needed before registering on the Kaiseki server, it takes time to make the validated data available. However, some researchers need data as soon as it is gathered in order to adjust their instruments during the experiments. To satisfy this requirement, a new visualization system has been under development. The new system has two ways to visualize the data as physical values from the raw data. If the conversion task is not complex, the NIFSscope, a visualization tool, converts the raw data into physics data by itself. If the task is too complex to handle, it asks the ANACalc server to make physics data. When the ANACalc server receives a request, it delegates calculation programs to convert the acquired data into physics data. Because the interfaces between the server and the calculation processes are independent of programming languages and operating systems, the calculation processes can be placed on different computers and the server load can be reduced. Therefore, the system can respond to changes in requirements by replacing the calculation programs, and can easily be expanded by increasing the number of calculation servers.