Method and apparatus for scheduling real-time and non-real-time access to a shared resource
AssigneeEquator Technologies, Inc.
InventorRudolf Johannes Bloks
Life of this patent
Estimating the term…
Abstract
A method and apparatus are provided in a computing environment for scheduling access to a resource. The method grants access to the resource by a non-real-time request when the non-real-time request can be completed before the latest possible start time at which a first real-time request must start service to timely complete all actual and anticipated real-time requests, otherwise granting the first real real-time request access to the resource.
Claims22
A scheduler comprising:
- a shared resource;
- a timer circuit to determine fixed required service completion times and non-fixed anticipated deadlines for access to the shared resource;
- and
- a control circuit to receive a plurality of received real-time requests and a plurality of anticipated real-time requests, the control circuit receiving a first of the plurality of received real-time requests subsequently determined by the timer circuit to have a fixed service completion time and to anticipate a first of the plurality of anticipated real-time requests subsequently determined by the timer circuit to have a non-fixed anticipated deadline, where the non-fixed anticipated deadline becomes a fixed required service completion time responsive to receiving a second of the plurality of received real-time requests.
A method comprising:
- sequentially receiving a plurality of received real-time requests;
- anticipating a plurality of anticipated real-time requests for access to a shared resource;
- determining a fixed required service completion time for a first of the plurality of received real-time requests;
- anticipating a non-fixed anticipated deadline for the first of the plurality of anticipated real-time requests;
- and
- updating the non-fixed anticipated deadline with a fixed required service completion time responsive to receiving a second of the plurality of received real-time requests.
A circuit for scheduling access to a shared memory resource in a multimedia computing environment that includes a processor operable to:
- receive a first received real-time request for access to a shared memory resource and a first non-real-time request for access to the shared memory resource, where the first received real-time request is part of a plurality of received real-time requests and the first non-real-time request is part of a plurality of non-real-time requests;
- anticipate receiving a first anticipated real-time request for access to the shared memory resource and store an anticipated deadline updated each clock cycle of the shared memory resource, where the first anticipated real-time request is part of a plurality of anticipated real-time requests;
- receive a second received real-time request for access to the shared memory resource, where the second received real-time request is part of the plurality of received real-time requests;
- replace the first anticipated real-time request anticipated deadline with a required service completion time responsive to the second received real-time request;
- and
- schedule the first non-real-time request if there is sufficient time for the shared memory resource to service the first non-real-time request, the first real-time request, and the second real-time request.
Dependencies are inferred from the claim text.