Abstract: An entry insertion apparatus (2000) inserts an entry into a node (3000) having a plurality of partial areas (3020) in which W (W being a positive integer equal to or greater than 2) or less entries are stored. Each entry can be searched for using an SIMD instruction to execute the same processing for the W entries in parallel. The entry insertion apparatus (2000) includes an insertion unit (2020) that inserts a new entry into the partial area (3020) in which the number of stored entries is 1 or more and less than W among the plurality of partial areas (3020), and a sorting unit (2040) that sorts the entries stored in the partial area (3020) when the number of entries stored in the partial area (3020) into which the new entry has been inserted is W.

Abstract: A database management device includes plural index blocks having a tree structure and each having an access counter and at least one index entry for identifying one piece of row data constituting a table data or another index block, and an access management device that updates the access counter of each of the plural index blocks accessed along the tree structure, the access being made in a manner such that the index entry is referred to in response to data manipulation to the table data.

Abstract: A database performance estimation device includes: a unit that switches at least two types of calculation methods according to configuration mode data and performs the switched calculation method to acquire a child block searching time (d) and a leaf block scanning time (s); a unit that switches at least two types of calculation methods according to the configuration mode data and performs the switched calculation method to acquire an entry inserting time (i); and a unit that calculates an average time required for identifying a target leaf block per one manipulation performed to the index block, an average time required for searching for a target entry in the identified leaf block, and an average time required for inserting a target entry in the identified leaf block, and estimates the total of the average times calculated above as an average processing time per one manipulation performed to the index block.

Abstract: A database management device includes: plural index blocks that have an index entry and an access counter including a read counter and a write counter; an access management unit that updates the read counter in response to a read access to an index block and updates the write counter in response to an update access to the index block; an extended block manipulation unit that acquires an extended block not identified with an index entry of another index block and serving as a destination of storing a new index entry on the basis of a result of comparison between the read counter and the write counter of a leaf block serving as a target of inserting the new index entry, and sets identification information for identifying the extended block to the leaf block serving as the target of insertion.

Abstract: An entry insertion apparatus (2000) inserts an entry into a node (3000) having a plurality of partial areas (3020) in which W (W being a positive integer equal to or greater than 2) or less entries are stored. Each entry can be searched for using an SIMD instruction to execute the same processing for the W entries in parallel. The entry insertion apparatus (2000) includes an insertion unit (2020) that inserts a new entry into the partial area (3020) in which the number of stored entries is 1 or more and less than W among the plurality of partial areas (3020), and a sorting unit (2040) that sorts the entries stored in the partial area (3020) when the number of entries stored in the partial area (3020) into which the new entry has been inserted is W.

Abstract: A data processing device has plural kinds of recording media and a data block management device. The data block management device classifies data blocks into plural groups and records each group on an appropriate recording medium. The data block management device has a memory unit, a group reconfiguration unit and a medium selection unit. Access trend information representing a trend of combinations of former groups and latter groups is stored in the memory unit. The group reconfiguration unit performs group reconfiguration processing by reference to the access trend information. Specifically, if a sequential access trend between two different groups is increased, they are integrated to generate a new group. If a sequential access trend within a certain group is decreased, the certain group is divided to generate a new group. The medium selection unit records the new group obtained as a result of the group reconfiguration on a corresponding recording medium.

Abstract: A database performance estimation device includes: a unit that switches at least two types of calculation methods according to configuration mode data and performs the switched calculation method to acquire a child block searching time (d) and a leaf block scanning time (s); a unit that switches at least two types of calculation methods according to the configuration mode data and performs the switched calculation method to acquire an entry inserting time (i); and a unit that calculates an average time required for identifying a target leaf block per one manipulation performed to the index block, an average time required for searching for a target entry in the identified leaf block, and an average time required for inserting a target entry in the identified leaf block, and estimates the total of the average times calculated above as an average processing time per one manipulation performed to the index block.

Abstract: A database management device includes: plural index blocks that have an index entry and an access counter including a read counter and a write counter; an access management unit that updates the read counter in response to a read access to an index block and updates the write counter in response to an update access to the index block; an extended block manipulation unit that acquires an extended block not identified with an index entry of another index block and serving as a destination of storing a new index entry on the basis of a result of comparison between the read counter and the write counter of a leaf block serving as a target of inserting the new index entry, and sets identification information for identifying the extended block to the leaf block serving as the target of insertion.

Abstract: A hierarchical load estimation system 1 has a memory device 2, an invocation count calculation module 100 and an issue count calculation module 200. Stored in the memory device 2 are a hierarchical invocation data 10 indicating design of hierarchical processing that is hierarchically invoked in a multi-layer system and an issuance data 20 specifying database manipulation sentences issued by database manipulation processing in the hierarchical processing. The invocation count calculation module 100 refers to the hierarchical invocation data 10 to calculate an invocation count of the database manipulation processing. The issue count calculation module 200 refers to the calculated invocation count and the issuance data 20 to calculate an issue count of database manipulation sentences issued by the database manipulation processing.

Abstract: A database management device includes plural index blocks having a tree structure and each having an access counter and at least one index entry for identifying one piece of row data constituting a table data or another index block, and an access management device that updates the access counter of each of the plural index blocks accessed along the tree structure, the access being made in a manner such that the index entry is referred to in response to data manipulation to the table data.

Abstract: A data processing system (1) includes an area allocating unit (102) which determines an area (202) of a nonvolatile recording medium (200) to be allocated, by a write period of periodic data to be written to the nonvolatile recording medium (200) and a valid period of the data, the nonvolatile recording medium (200) permanently storing data; an additionally writing unit (104) which accepts the data to be written to the nonvolatile recording medium (200), and temporarily stores the data in a volatile recording medium (300); and a permanent storage implementation unit (106) which writes the data temporarily stored in the volatile recording medium (300), to the area (202) of the nonvolatile recording medium (200) determined by the area allocating unit (102).

Abstract: A computer system is provided with an event counter, a CPU, a memory, an external device, a hub M31 and a hub I33. The computer system is further provided with a clock change module 50. System software 60 and applications 70a to 70m operate, and the clock change module 50 specifies a clock with which the applications 70 are executed. Based on stall cycles read from the event counter 21, the clock change module 50 specifies the clock that maximizes performance or power consumption characteristic while satisfying execution constraint 90.

Abstract: A data processing device has plural kinds of recording media and a data block management device. The data block management device classifies data blocks into plural groups and records each group on an appropriate recording medium. The data block management device has a memory unit, a group reconfiguration unit and a medium selection unit. Access trend information representing a trend of combinations of former groups and latter groups is stored in the memory unit. The group reconfiguration unit performs group reconfiguration processing by reference to the access trend information. Specifically, if a sequential access trend between two different groups is increased, they are integrated to generate a new group. If a sequential access trend within a certain group is decreased, the certain group is divided to generate a new group. The medium selection unit records the new group obtained as a result of the group reconfiguration on a corresponding recording medium.

Abstract: A computer system is provided with an event counter, a CPU, a memory, an external device, a hub M31 and a hub I33. The computer system is further provided with a clock change module 50. System software 60 and applications 70a to 70m operate, and the clock change module 50 specifies a clock with which the applications 70 are executed. Based on stall cycles read from the event counter 21, the clock change module 50 specifies the clock that maximizes performance or power consumption characteristic while satisfying execution constraint 90.

Abstract: A hierarchical load estimation system 1 has a memory device 2, an invocation count calculation module 100 and an issue count calculation module 200. Stored in the memory device 2 are a hierarchical invocation data 10 indicating design of hierarchical processing that is hierarchically invoked in a multi-layer system and an issuance data 20 specifying database manipulation sentences issued by database manipulation processing in the hierarchical processing. The invocation count calculation module 100 refers to the hierarchical invocation data 10 to calculate an invocation count of the database manipulation processing. The issue count calculation module 200 refers to the calculated invocation count and the issuance data 20 to calculate an issue count of database manipulation sentences issued by the database manipulation processing.

Abstract: A load estimating system which estimates a load of a database, includes a storage unit, a record number calculating module, and an access number calculating module. The storage unit includes structure information defining a structure of a table in said database, first distribution information indicating probability distribution of condition variables of a query for said table, and second distribution information indicating probability distribution of values in said table. The record number calculating module calculates the average number of selected records which meet conditions defined by-condition variables, based on said structure information, said first distribution information and said second distribution information. The access number calculating module calculates the number of block accesses for said query based on the average number of selected records.

Abstract: A technique for supporting the sizing of hierarchical computer resources is provided. The sizing support system includes a memory, a table definition module and field calculation module. The memory stores lower level resource data which indicates the kinds and the maximum number of lower level resources belonging to the resources of each hierarchy level, and resource data which indicates parameters of resources of each hierarchy level. The processing is executed in a unit of three levels. In the unit process of three levels, the table definition module generates a table frame indicating the possible configurations of first resource by referring the lower level resource data. The field calculation module calculates the performance and cost of the configuration in the table frame by referring the resource data.

Abstract: A load estimating system which estimates a load of a database, includes a storage unit, a record number calculating module, and an access number calculating module. The storage unit includes structure information defining a structure of a table in said database, first distribution information indicating probability distribution of condition variables of a query for said table, and second distribution information indicating probability distribution of values in said table. The record number calculating module calculates the average number of selected records which meet conditions defined by-condition variables, based on said structure information, said first distribution information and said second distribution information. The access number calculating module calculates the number of block accesses for said query based on the average number of selected records.

Abstract: A technique for supporting the sizing of hierarchical computer resources is provided. The sizing support system includes a memory, a table definition module and field calculation module. The memory stores lower level resource data which indicates the kinds and the maximum number of lower level resources belonging to the resources of each hierarchy level, and resource data which indicates parameters of resources of each hierarchy level. The processing is executed in a unit of three levels. In the unit process of three levels, the table definition module generates a table frame indicating the possible configurations of first resource by referring the lower level resource data. The field calculation module calculates the performance and cost of the configuration in the table frame by referring the resource data.

Abstract: A data grouping means divides data items stored in a cache memory section into groups of data having different access patterns. The priority assigning means assigns an order of priorities to data items in each group that the priority assigning means manages according to an individual caching algorithm. The lowest priority determining means determines the lowest priority group when there is not enough unused memory space in the cache memory section and it is necessary to purge a data item. The data operating means purges the lowest priority data in the lowest priority group. Thus the groups of data having different access patterns can be cached effectively.