The advanced data searching system with AMGA at the Belle II experiment

We have developed a metadata service for the Belle experiment which provides a mechanism to locate files using descriptive information. However, for the Belle II experiment, we will have 50–60 times more data. This metadata service may have problems with performance, scalability, and durability when employed at Belle II. These issues are compounded when metadata searches are extended to the event-level. Accordingly, we have designed a new metadata scheme for Belle II which significantly reduces disk space and propose a new metadata service which provides good performance and scalability based on AMGA (Arda Metadata catalog for Grid Application). We find that the use of event-level metadata provides an efficient scheme for processing events with many tracks.