Belle II is an upgraded detector to allow the experiment to record the enormous numbers of particle processes that are produced by the SuperKEKB accelerator. The Belle II experiment will accumulate 50 times more data than the previous experiment, Belle, and pursue violations of the symmetry between particles and anti-particles as well as new laws of physics.
The construction of the SuperKEKB accelerator, which is an upgrade of the KEKB accelerator that operated for a decade from 1999, started in 2010 with the goal of 40 times higher collision performance. The energies of the electron and positron beams were changed to 7 and 4 billion electron volts, respectively. The expected collision performance is achieved by squeezing beam sizes to the nanometer level. The accelerator plans to start operation in JFY 2014. In order to cope with the higher beam intensity, the detector is also being upgraded. The inner detectors will be replaced with new ones to improve performance. A new collaboration, Belle II, has formed based on the Belle collaboration; it now consists of about 600 researchers from about 100 universities/laboratories in 23 countries/areas (as of December 2013). The researchers work together on construction of the detector, acquisition of data, and physics analyses. Analyses on 50 times more data than available to the Belle experiment will reveal violations of the symmetry between particles and anti-particles, rare phenomena that occurred in the early stages of the universe, and the nature of unknown particles and new interactions. These may lead to new laws of physics and an understanding of why anti-matter has disappeared from the universe.
Preamble: The central computing system at KEK is replaced every four years. The main user must be Belle II until the data taking ends (in 2024).
Belle : mDST (necessary for physics analysis) is stored on the disk as well as the tape library. The data is still frequently read by active analysis users. All data will be preserved by migrating to the next system. We experienced data loss in the previous data migration. Main causes of this trouble were the short migration period, miscommunication between researchers and operators and the lack of the validation scheme after the migration. We will improve the process of the future migration.
Belle : raw data (1PB) and other format (incl. simulation, ~1.5PB) are stored at the KEK central computing system. This data will be migrated to the next system, at least (data will be preserved until 2020). However, there is no plan thereafter, because the data will be superseded by Belle II. And a full set of mDST was copied at PNNL in USA.
Belle II : data taking is not started yet. But raw data will be stored at KEK and another set will be copied in some places outside Japan. Also, the replicas of the mDST will be distributed to the world-wide collaborated computing sites.
Belle : all documentation is stored in the local web server and INDICO system. They are still active and accessible, but not well catalogued at all.
Belle II : Using twiki, invenio, svn and INDICO system.
Belle : software has been fixed since 2009 except for some patches. The baseline of the OS is still SL5, but it was migrated to SL6. In parallel, the Belle data I/O tool is developed and integrated in the Belle II software. Thanks to this, the Belle data can be analysed under the Belle II software environment. Other Belle handmade analysis tools are being integrated, too. Software version is maintained with SVN.
Belle II : basic features witch are necessary for the coming data taking have been implemented. But need more tuning and improvement. The software version is maintained by SVN. SL5/6 32/64-bits, Ubuntu 14.02 LTS are supported
Continued analyses by Belle.
Belle and Belle II
Quantitative measures (# papers, PhDs etc) exist
Belle : During the data taking period (1999-2010), averaged number of journal publications is ~30 papers/year and the number of PhD is ~12/year.
After the data taking, a moderate decreasing tendency can be seen, but the analysis is still active. (~20 publications/year and ~7 PhDs/year).
Belle : Comparing with the data from Hadron colliders, the Belle data has advantage to analyse the physics modes with missing energy and neutral particles. Until the Belle II starts, these data are unique as well as BaBar’s data.
Belle II : Belle data will be superseded by 2020. After that, the data must be unique samples.
Belle / Belle II : in some stage, the Belle data must be treated as a part of the Belle II data. And resources for the Belle data will be included in the Belle II computing/human resources.
Construction of the Belle II detector/SuperKEKB accelerator as well as the Belle II distributed computing system.
A couple of items have to be implemented in the Belle II software framework to analyse the Belle data. Further check for the performance and reproducibility is also necessary.
Expect to be able to analyse the Belle data within the Belle II software framework. It provides us the less human resource to maintain the Belle software, and the longer lifetime of the Belle data analysis.