Abstract: An information processing system includes a plurality of PMM and data transmission paths for connection between the PMM and transmitting a value of a PMM to another PMM. A memory of each PMM holds a list of values of first items arranged in the ascending order or descending order without overlap and/or a list of values of the second item to be shared. A memory module of each PMM transmits a value contained in the value list to another PMM, receives a value contained in the value list from the another PMM, references the value list of the first item and the value list of the second item of the another PMM, and generates a list of common values considering the values contained in the value lists of the first item and the second item of all the other PMM.

Abstract: Large-scale table data stored in a shared memory are sorted by a plurality of processors in parallel. According to the present invention, the records subjected to processing are first divided for allocation to the plurality of processors. Then, each processor counts the numbers of local occurrences of the field value sequence numbers associated with the records to be processed. The numbers of local occurrences of the field value sequence numbers counted by each processor is then converted into global cumulative numbers, i.e., the cumulative numbers used in common by the plurality of processors. Finally, each processor utilizes the global cumulative numbers as pointers to rearrange the order of the allocated records.

Abstract: An ascending ordered list without duplication is generated based on a value list divided and held by multiple memory modules. An information processing system has multiple PMMs (Processor Memory Modules), and the PMMs are interconnected via a data transmission path. The memory in the PMM has a list of values, which are ordered in ascending or descending order without duplication. The PMM determines, for a storage value in the value list (LOCAL_LIST) held by the PMM, whether or not the memory module is a representative module representing one or more memory modules holding the storage value based on rankings determined for the individual PMMs and the value lists received from the other PMMs, and if the memory module is determined to be the representative module (RV-0 . . . RV-7), associates to the storage value and stores information indicating that the memory module is the representative module.

Abstract: A relationship among data in a tree data structure is effectively traced by assigning nodes respective unique node identifiers, a parent-child relationship is represented by a C->P array containing node identifiers for parent nodes associated with node identifiers for non-root nodes that are nodes other than a root node, a list of vertex nodes is provided by storing node identifiers for respective vertex nodes in order to represent one or more node groups, each node group including a particular node, which is the vertex node, and descendant nodes of the particular node, and the node identifier for the vertex node belonging to other vertex nodes are deleted from the list of vertex nodes by referring to the C->P array such that neither vertex nodes nor respective descendant nodes of the vertex nodes duplicate each other across vertex nodes identified by the node identifiers included in the list of vertex nodes.

Abstract: A searching method for tree data structure topology employs an array generated as a search key array representing a tree serving as search key to represent a parent-child relationship between nodes. The array has the node identifier assigned to a parent node associated with a corresponding non-root node. Non-root nodes are nodes other than a root node, and the node identifier assigned to the parent node corresponds to the node identifier assigned to associated non-root nodes. An offset is applied to node identifiers in the search key array, the node identifiers to which offset is applied are compared with corresponding node identifiers contained in a portion in a C-P array, the portion determined according to the offset, and a vertex node list indicating a search result is stored, wherein the search result contains node identifiers included in a portion of the C-P array whose all nodes matched in comparison.

Abstract: Large-scale table format data on a common memory is sorted in parallel by a plurality of processors. A record to be processed is divided and assigned at first to the processors. Next, each processor counts the number of local appearances of item value numbers related to the record to be processed. Next, the number of local appearances of the item value numbers counted at the processors is converted into the global accumulation of item value numbers, i.e., the accumulation to be commonly used between the processors. Finally, each processor utilizes the global accumulation as a pointer to change the order of the assigned records.

Abstract: Table-formatted data is divided into information blocks, each including a value list where item values are arranged in order of item value number corresponding to the item value of each item, and a pointer array where pointer values, each indicate an item value number, are arranged in order of unique record number. A part of records is selected as a subset, and in response to the selection, an ordered set array including the record numbers of the subset is created. Each record number in the ordered set array indicates the position of one of the pointer values in the pointer array, which are arranged in the item value number array correspondingly to positions where the record numbers are arrayed. With reference to values in the item value number array, a second pointer array wherein position elements are arranged, and a second value list wherein value elements are arranged, are generated.

Abstract: In parallel computers, sorting and calculation of large-scale data are realized while large-scale data is held in the respective processors without sharing the large-scale data between the processors so as to reduce communication between the processors. An information processing method gives global dimension value numbers common to all the processing modules to the dimension values for calculation, calculates measures for each of the dimension value numbers within the processing module, and lastly calculates measures commonly between all processing modules. The value list and pointer arrangement to the value list are locally held in each processing module and the order of the dimension values as a reference is globally held between processing modules. As a result, it is possible to eliminate mutual access by processing modules for acquiring data required for calculation and only data required for deciding the order of the dimension values is communicated between the processing modules.

Abstract: Large-scale table data stored in a shared memory are sorted by a plurality of processors in parallel. According to the present invention, the records subjected to processing are first divided for allocation to the plurality of processors. Then, each processor counts the numbers of local occurrences of the field value sequence numbers associated with the records to be processed. The numbers of local occurrences of the field value sequence numbers counted by each processor is then converted into global cumulative numbers, i.e., the cumulative numbers used in common by the plurality of processors. Finally, each processor utilizes the global cumulative numbers as pointers to rearrange the order of the allocated records.

Abstract: It is possible to express a tree-type data structure so as to effectively trace the relationship between data in the tree-type data structure (for example, parent-child, ancestor, descendant, brothers, generations). In the tree-type data structure, for each of non root nodes which are the nodes excluding the root node, their parent nodes are correlated so that the parent-child relationship between the nodes is expressed by using “child->parent” relationship. Accordingly, by specifying a child node, it is possible to promptly specify the only one parent node corresponding to the child node.

Abstract: Large-scale table format data on a common memory is sorted in parallel by a plurality of processors. A record to be processed is divided and assigned at first to the processors. Next, each processor counts the number of local appearances of item value numbers related to the record to be processed. Next, the number of local appearances of the item value numbers counted at the processors is converted into the global accumulation of item value numbers, i.e., the accumulation to be commonly used between the processors. Finally, each processor utilizes the global accumulation as a pointer to change the order of the assigned records.

Abstract: Concatenated join tables are converted into a tree structure at a high speed. First, table-formatted data is divided into one or more information blocks consisting of a value list containing item values in the sequence of the item value numbers corresponding to item values belonging to particular items and a pointer sequence containing pointer values for indicating the item value numbers in the sequence of the unique record number. Next, table-formatted data whose items are to be made common are selected, value lists of a predetermined item are made equivalent, and join is executed. Join is repeated between necessary table-formatted data. Next, among the concatenated table data, one which is to be a root is selected. According to the table-formatted data concatenation, depth of the table-formatted data is decided. According to the table-formatted data concatenation and depth, a value indicating a record is arranged on a tree description table describing the tree.

Abstract: A data compiling method for conversion into the form of an information block containing a value list in which table format data expressed as an array of records including items and item values contained in the items are so arranged that the item values are arranged in a predetermined order without any redundancy, and a position designation array including position designation numbers designating the item values in the value list and related to the record numbers. Adjoining partial intermediate blocks (811-0, 811-1) which include an item value array having item values related to the record numbers, an order designation array (VLP) for designating the positions of the item values in the item value array, and a position designating array (PV) for designating elements in the order designation array are merged to create a new partial intermediate block (812-0), and the operation of merging the partial intermediate blocks is repeated till the blocks are merged into one.

Abstract: A computer system provides distributed memory computer architecture achieving extremely high speed parallel processing, and includes: a CPU modules, a plurality of memory modules, each module having a processor and RAM core, and a plurality of sets of buses making connections between the CPU and the memory modules and/or connections among memory modules, so the various memory modules operate on an instruction given by the CPU. A series of data having a stipulated relationship is given a space ID and each memory module manages a table containing at least the space ID, the logical address of the portion of the series of data managed, the size of the portion and the size of the series of data, and, the processor of each memory module determines if the portion of the series of data managed is involved in a received instruction and performs processing on data stored in the RAM core.

Abstract: A plurality of table-formatted data is joined by dividing each table-formatted data into one or more information blocks consisting of a value list containing item values in the order of item value numbers corresponding to item values belonging to a particular item and a pointer array containing pointer values specifying item value numbers in the order of uniquely-defined record number. When joining a plurality of table formats, equivalent items between the table-formatted data are found and value lists contained in the information block concerning the equivalent items are compared and the two value lists are made equivalent, then data of one of the table formats is sorted according to a key item and using the array obtained in this process, calculation is performed to obtain addresses indicating storage positions of record numbers of table-formatted data and other table-formatted data. Using the calculated addresses, joined table-formatted data is displayed.

Abstract: Concatenated join tables are converted into a tree structure at a high speed. First, table-formatted data is divided into one or more information blocks consisting of a value list containing item values in the sequence of the item value numbers corresponding to item values belonging to particular items and a pointer sequence containing pointer values for indicating the item value numbers in the sequence of the unique record number. Next, table-formatted data whose items are to be made common are selected, value lists of a predetermined item are made equivalent, and join is executed. Join is repeated between necessary table-formatted data. Next, among the concatenated table data, one which is to be a root is selected. According to the table-formatted data concatenation, depth of the table-formatted data is decided. According to the table-formatted data concatenation and depth, a value indicating a record is arranged on a tree description table describing the tree.

Abstract: A method to perform the insertion, deletion and updating of data in table-format data quickly and appropriately. A CPU 12 accepts a record number as a subscript, generates a subscript conversion array for giving an offset value corresponding to the range of the subscript in question, identifies the insertion position which indicates the position of the field value to be inserted, and, in the subscript conversion array, gives an offset value that defines the range of the corresponding subscript and also identifies the end of the array, and in the subscript conversion array, gives an offset value that increments the corresponding range of subscripts and also decrements the accepted subscript, and places the field value to be inserted at the stipulated end position, such that an offset value according to the range of subscripts within the subscript conversion array is given as the subscript.

Abstract: A plurality of table-formatted data can be joined at a high speed without requiring use of a large capacity of memory. Each table-formatted data is divided into one or more information blocks consisting of a value list containing item values in the order of item value numbers corresponding to the item values belonging to a particular item and a pointer array containing pointer values specifying the item value numbers in the order of uniquely-defined record number. When joining a plurality of table formats, equivalent items between the table-formatted data are found and the value lists contained in the information block concerning the equivalent items are compared and the two value lists are made equivalent. After this, data of one of the table formats is sorted according to a key item and by using the array obtained in this process, calculation is performed to obtain addresses indicating storage positions of record numbers of the table-formatted data and the other table-formatted data.

Abstract: The object of this invention is to perform the sorting, compiling and joining of data at extremely high speeds. This is achieved by a distributed memory type information processing system comprising: a CPU module, a plurality of memory modules, each of which having a processor and RAM core, and a plurality of sets of buses that make connections between the CPU and memory modules and/or connections among memory modules, where the processors of the various memory modules execute the processing of arrays managed by the one or more memory modules based on instructions given by the CPU to the processors of the various memory modules.

Abstract: This invention provides a method and apparatus for searching for and tabulating table-format data that not only has the functions of a conventional data table but also greatly increases the speed of searching for and tabulating large amounts of data.