Abstract: In a distribution system including a plurality of physical machines, when an output file output by first processing performs multistage processing input to second processing, a physical machine which is a deployment destination of a first virtual machine for performing the first processing, a physical machine which is a storage destination of each data block included in the output file, and a physical machine which is a deployment destination of a second virtual machine for performing the second processing are determined based on relationship between the first and second processing.

Abstract: A file on file storage is archived as an object on object storage. In this case, a correspondence between UUID which is an identifier of the object and a path or an inode number which is an identifier of the file is recorded. This is referred to as correlation between the file and the object. However, because of changing of the path due to the renaming of the file, or changing of the inode number due to overwriting of the file, the correlation is likely to be lost. The invention provides a technique that restores the lost correlation and causes the correlation between the file and the object to be maintained.

Abstract: In a distribution system including a plurality of physical machines, when an output file output by first processing performs multistage processing input to second processing, a physical machine which is a deployment destination of a first virtual machine for performing the first processing, a physical machine which is a storage destination of each data block included in the output file, and a physical machine which is a deployment destination of a second virtual machine for performing the second processing are determined based on relationship between the first and second processing.

Abstract: A file on file storage is archived as an object on object storage. In this case, a correspondence between UUID which is an identifier of the object and a path or an inode number which is an identifier of the file is recorded. This is referred to as correlation between the file and the object. However, because of changing of the path due to the renaming of the file, or changing of the inode number due to overwriting of the file, the correlation is likely to be lost. The invention provides a technique that restores the lost correlation and causes the correlation between the file and the object to be maintained.

Abstract: Provided is a method for retaining metadata inheritance relationship in an object storage and for interpreting the relationship. A file and directory inheriting metadata of an upper-level directory allocates the UUID of an object corresponding to the inheritance source directory to metadata of an object of a CAS device corresponding to the file and directory as inheritance information. The inheritance information stored in the CAS device is interpreted in a metadata conversion apparatus (see FIG. 1).

Abstract: A search system using a table search server and a file search server as transmission destination candidates for search queries, wherein search speed is assumed to be higher for a search in the form of file data than for table data, the table data is converted to file data and stored in the file search server. Created are a search query history management table for accumulating and depositing search query history, and a characteristic determination rule management table for managing the rules of determining that the search speed is higher for a search made in the form of file data than for table data. The search system applies the characteristic determination rules to the search query history and specifies the table data. The search system acquires the specified table data from the table search server, converts the data to file data, and stores the data in the file search server.