SCAN OPTIMIZATION USING BLOOM FILTER SYNOPSIS

    公开(公告)号:CA2876466A1

    公开(公告)日:2016-06-29

    申请号:CA2876466

    申请日:2014-12-29

    Abstract: An illustrative embodiment for optimizing scans using a Bloom filter synopsis, defines metadata to encode distinct values in a range of values associated with a particular portion of a managed object in a database management system into a probabilistic data structure of a Bloom filter that stores an indicator, encoded in a fixed size bit map with one or more bits, indicating whether an element of the particular portion of the managed object is a member of a set of values summarized in the Bloom filter using a value of / or definitely not in the set using a value of 0. The Bloom filter is compressed to create a compressed Bloom filter. The Bloom filter is added to the metadata associated with the managed object and used when testing for values associated with predicates.

    MEMORY MANAGEMENT IN PRESENCE OF ASYMMETRICAL MEMORY TRANSFER COSTS

    公开(公告)号:CA2876379A1

    公开(公告)日:2016-06-29

    申请号:CA2876379

    申请日:2014-12-29

    Abstract: An illustrative embodiment tunes memory in presence of asymmetrical memory transfer costs by assigning a threshold value asynchronously to a memory consumer and a bias value to the threshold. Responsive to an ability to free memory by the consumer and to reclaim memory, direct the memory consumer to reclaim memory. Responsive to a time out or the reclaim exceeds a predetermined amount of time for the memory consumer, multiply the biased threshold value by a predetermined adjustment factor. Responsive to freeing memory under the predetermined time, divide the biased benefit threshold value by the predetermined adjustment factor. Update per memory consumer control values and logically partition the memory according to an ability of the memory consumer to free the memory in a defined ratio. The memory consumer reclaims memory from the second portion and replenishes the second portion asynchronously from the first portion, maintaining the defined ratio.

    SCAN OPTIMIZATION USING BLOOM FILTER SYNOPSIS

    公开(公告)号:CA2876466C

    公开(公告)日:2022-07-05

    申请号:CA2876466

    申请日:2014-12-29

    Abstract: An illustrative embodiment for optimizing scans using a Bloom filter synopsis, defines metadata to encode distinct values in a range of values associated with a particular portion of a managed object in a database management system into a probabilistic data structure of a Bloom filter that stores an indicator, encoded in a fixed size bit map with one or more bits, indicating whether an element of the particular portion of the managed object is a member of a set of values summarized in the Bloom filter using a value of / or definitely not in the set using a value of 0. The Bloom filter is compressed to create a compressed Bloom filter. The Bloom filter is added to the metadata associated with the managed object and used when testing for values associated with predicates.

    AUTONOMIC REALLOCATION OF MEMORY AMONG BUFFER POOLS

    公开(公告)号:CA2408481A1

    公开(公告)日:2004-04-17

    申请号:CA2408481

    申请日:2002-10-17

    Applicant: IBM CANADA

    Abstract: Buffer pools permit quick access to data. A simulated buffer pool extensi on (SBPX) is created for each buffer pool in a set of buffer pools. Data victimized from a buffer pool is represented in the associated SBPX. Requests for data that is not resident in a buffer pool but is represented in the associated SBPX are tallied. Periodically, an expected efficiency benefit of increasing the capacity of each buffer pool i s determined from the tallies. Memory is reallocated from the buffer pool with the lowest expected efficiency benefit having remaining reallocatable memory to the buffer pool with the highest expected efficiency benefit having remaining reallocatable memory, until either one or both of the buffer pools exhausts its reallocatable memory. This is repeated until all reallocatable memory has been reallocated, until only one buffer pool wi th reallocatable memory remains, or until all buffer pools with remaining reallocatable memor y have substantially the same expected efficiency benefit.

    Schnelle Prädikattabellensuchen unter Verwendung einer Architektur mit einer Einzelanweisung und mehrfachen Daten

    公开(公告)号:DE102013204515A1

    公开(公告)日:2013-10-02

    申请号:DE102013204515

    申请日:2013-03-15

    Applicant: IBM

    Abstract: Bereitgestellt wird ein Ansatz, bei dem ein Prozessor eine Suchanforderung zum Suchen von in einer Datentabelle enthaltenen Daten empfängt. Der Prozessor wählt eine Spalte in der Datentabelle aus, die der Suchanforderung entspricht, und ruft Spaltendateneinträge aus der ausgewählten Spalte ab. Außerdem ermittelt der Prozessor die Breite der ausgewählten Spalte und wählt einen Suchalgorithmus auf der Grundlage der ermittelten Spaltenbreite aus. Danach lädt der Prozessor die Spaltendateneinträge in Spaltendatenvektoren und berechnet Suchergebnisse aus den Spaltendatenvektoren unter Verwendung des ausgewählten Suchalgorithmus.

    AUTONOMIC REALLOCATION OF MEMORY AMONG BUFFER POOLS

    公开(公告)号:CA2408481C

    公开(公告)日:2009-10-06

    申请号:CA2408481

    申请日:2002-10-17

    Applicant: IBM CANADA

    Abstract: Buffer pools permit quick access to data. A simulated buffer pool extensi on (SBPX) is created for each buffer pool in a set of buffer pools. Data victimized from a buffer pool is represented in the associated SBPX. Requests for data that is not resident in a buffer pool but is represented in the associated SBPX are tallied. Periodically, an expected efficiency benefit of increasing the capacity of each buffer pool i s determined from the tallies. Memory is reallocated from the buffer pool with the lowest expected efficiency benefit having remaining reallocatable memory to the buffer pool with the highest expected efficiency benefit having remaining reallocatable memory, until either one or both of the buffer pools exhausts its reallocatable memory. This is repeated until all reallocatable memory has been reallocated, until only one buffer pool wi th reallocatable memory remains, or until all buffer pools with remaining reallocatable memor y have substantially the same expected efficiency benefit.

    Effiziente Durchführung von Einfüge- und Punktabfrage-Operationen in einem Spaltenspeicher

    公开(公告)号:DE112016000776T5

    公开(公告)日:2017-10-26

    申请号:DE112016000776

    申请日:2016-02-04

    Applicant: IBM

    Abstract: Ein Verfahren beinhaltet ein logisches Organisieren von Datenobjekten in einer ersten Hierarchie durch einen Objekthierarchieprozessor. Ein Teil der Datenobjekte in der ersten Hierarchie beinhaltet logisch Gruppierungen von anderen Datenobjekten. Der Objekthierarchieprozessor organisiert die Datenobjekte physisch zwischen zwei oder mehr Arten von Arbeitsspeichern in einer zweiten Hierarchie. Ein weiterer Teil der Datenobjekte in der zweiten Hierarchie beinhaltet physisch Gruppierungen von anderen Datenobjekten. Gruppierungen der Datenobjekte in der zweiten Hierarchie werden dynamisch zwischen den zwei oder mehr Arten von Arbeitsspeichern verschoben. Zugriffsebenen der Datenobjekte werden anhand einer Datenstruktur erfasst, die Gruppierungen der Datenobjekte in der ersten Hierarchie auf Metadaten-Informationen wie z. B. kombinierte Zugriffshäufigkeiten der Datenobjekte und eine momentane Anzahl von Zugreifern auf die Datenobjekte in jeder Gruppierung der Datenobjekte abbildet.

Patent Agency Ranking