Need to retrieve the HPC utilization statistics in realtime. In the current version, we may experience the performance degradation. If we store the following information into separate dataset, I expect to avoid the degradation. I think the HPC executor is one of the candidates for extensions.

• userID: email
• jobID: Job ID assigned by local scheduler, e.g., load leveler
• solver name
• start time : the time when the job is running on computing resources
• end time : the time when the job is finished on computing resources
• queued duration: the duration of waiting in the queue on computing resources
• job status : the final status of job
• duration : = (end time - start time)
• model file size, e.g., mesh file in CFD simulation
• result file size