System and methods for optimizing data transfer among various resources in a distributed environment

System providing methodology for optimizing data transfer in a distributed environment is described. In one embodiment, for example, in a distributed shared disk cluster environment comprises a plurality of nodes, each node storing database pages in different size buffer pools configured for local a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: BHATT VADIRAJA P, PASUPULETI KANTIKIRAN KRISHNA, NALAWADE VAIBHAV A
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:System providing methodology for optimizing data transfer in a distributed environment is described. In one embodiment, for example, in a distributed shared disk cluster environment comprises a plurality of nodes, each node storing database pages in different size buffer pools configured for local access and configured independent from other nodes in the cluster, the database pages residing at any buffer pool across nodes without requiring that a page be present at an identical buffer pool at all nodes in the cluster, a method of the present invention for transferring database pages among nodes comprises: generating a request at a first node for a set of database pages, the request including a timestamp for determining what are latest versions of the database pages; in response to the request, determining where latest versions of the database pages of the set reside among the plurality of nodes without regard to what size of buffer pool any given database page may reside in; if the database pages of the set reside among the plurality of nodes, initiating transfer of particular database pages to the first node as required for providing the first node with the latest versions of the database pages; and, otherwise if the database pages of the set do not reside among the plurality of nodes, instructing the first node to fetch the database pages from disk.