Directory-based prediction methods and apparatus for shared-memory multiprocessor systems
AssigneeAgere Systems Inc.
InventorsStefanos Kaxiras, Reginald Clifford Young
Life of this patent
Estimating the term…
Abstract
A set of predicted readers are determined for a data block subject to a write request in a shared-memory multiprocessor system by first determining a current set of readers of the data block, and then generating the set of predicted readers based on the current set of readers and at least one additional set of readers representative of at least a portion of a global history of a directory associated with the data block. In one possible implementation, the set of predicted readers are generated by applying a function to the current set of readers and one or more additional sets of readers.
Claims34
A method of determining a set of predicted readers of a data block in a multiprocessor system, the method comprising the steps of:
- determining a current set of readers of a data block which is subject to a write request;
- and
- generating the set of predicted readers based on the current set of readers and at least one additional set of readers representative of at least a portion of a global history of a directory associated with the data block.
An apparatus for determining a set of predicted readers of a data block in a multiprocessor system, the apparatus comprising:
- a processor node operative to determine a current set of readers of a data block which is subject to a write request, and to implement a prediction mechanism which generates the set of predicted readers based on the current set of readers and at least one additional set of readers representative of at least a portion of a global history of a directory associated with the data block.
A multiprocessor system comprising:
- a plurality of processor nodes, at least a given one of the processor nodes being operative to determine a current set of readers of a data block which is subject to a write request, the given processor node implementing a prediction mechanism which generates a set of predicted readers of the data block based on the current set of readers and at least one additional set of readers representative of at least a portion of a global history of a directory associated with the data block.
Dependencies are inferred from the claim text.