Dynamic allocation of platform resources
A dynamic allocator for providing platform resource candidates is disclosed. In an implementation, a platform resource allocator receives a request from a workload initiator such as, an application, for a platform resource recommendation. The platform resource allocator analyzes performance capabilities and utilization metrics of a plurality of platform resources for each of a plurality of resource. The plurality of platform resources includes one or more graphics processor units (GPUs) and one or more accelerated processing units (APUs). The platform resource allocator dynamically provides the platform resource recommendation to the workload initiator to select one or more of the plurality of platform resources to execute a workload based on the performance capabilities and utilization metrics.
1 . A method of providing platform resource candidates, the method comprising:
identifying one or more platform resource candidates from a plurality of platform resources based on a request received from a workload initiator for platform resource candidates, performance capabilities of the plurality of platform resources, utilization metrics of the plurality of platform resources, and non-discoverable performance capabilities that are non-discoverable by an operating system, wherein the plurality of platform resources includes one or more graphics processor units (GPUs) and one or more accelerated processing units (APUs); and
dynamically providing a recommendation including the identified one or more platform resource candidates to the workload initiator.
2 . The method of claim 1 , wherein the platform resource candidates are also identified based on non-performance capabilities and performance limitations of the plurality of platform resources.
3 . The method of claim 1 , further comprising collecting the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations for the plurality of platform resources in a platform resources library.
4 . The method of claim 1 , further comprising consulting a platform resources library to determine the non-discoverable performance capabilities that are non-discoverable by an operating system, performance limitations, and historical performance data for the plurality of platform resources based on the request from the workload initiator.
5 . The method of claim 1 , further comprising monitoring the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations for the plurality of platform resources during runtime of the workload.
6 . The method of claim 1 , further comprising dynamically providing, during runtime of the workload, updated platform resource candidates based on adjusted performance capabilities and adjusted utilization metrics of the plurality of platform resources.
7 . The method of claim 1 , further comprising ranking each of the plurality of platform resources based on the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations.
8 . The method of claim 1 , further comprising predicting a utilization impact to execute the workload for each of the plurality of platform resources based on the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations.
9 . The method of claim 1 , wherein the performance capabilities include operational modes of the plurality of platform resources, interconnection capabilities between the plurality of platform resources, power capabilities of the plurality of platform resources, and peripheral device capabilities.
10 . The method of claim 1 , wherein the plurality of platform resources further includes a local memory, a system memory, one or more peripheral devices, a display controller, a bus interface controller, and a memory subsystem controller, wherein at least one of the one or more GPUs is a discrete GPU.
11 . An apparatus of providing platform resource candidates, the apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed therein computer program instructions that, when executed by the computer processor, cause the apparatus to:
identify one or more platform resource candidates from a plurality of platform resources based on a request received from a workload initiator for platform resource candidates, performance capabilities and utilization metrics of the plurality of platform resources, and non-discoverable performance capabilities that are non-discoverable by an operating system, wherein the plurality of platform resources includes one or more graphics processor units (GPUs) and one or more accelerated processing units (APUs); and
dynamically provide a recommendation including the identified one or more platform resource candidates to the workload initiator.
12 . The apparatus of claim 11 , further comprising computer program instructions that, when executed, cause the apparatus to consult a platform resources library to determine the non-discoverable performance capabilities that are non-discoverable by an operating system, performance limitations, and historical performance data for the plurality of platform resources based on the request from the workload initiator.
13 . The apparatus of claim 11 , further comprising computer program instructions that, when executed, cause the apparatus to monitor the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations for the plurality of platform resources during runtime of the workload.
14 . The apparatus of claim 11 , further comprising computer program instructions that, when executed, cause the apparatus to dynamically provide, during runtime of the workload, updated platform resource candidates based on adjusted performance capabilities and adjusted utilization metrics of the plurality of platform resources.
15 . The apparatus of claim 11 , further comprising computer program instructions that, when executed, cause the apparatus to rank each of the plurality of platform resources based on the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations.
16 . The apparatus of claim 11 , further comprising computer program instructions that, when executed, cause the apparatus to predict a utilization impact to execute the workload for each of the plurality of platform resources based on the performance capabilities, the utilization metrics, the non-discoverable performance capabilities that are non-discoverable by an operating system, and performance limitations.
17 . A computer program product of providing platform resource candidates, the computer program product disposed upon a non-transitory computer readable medium, the computer program product comprising computer program instructions that, when executed, cause a computer to:
identify one or more platform resource candidates from a plurality of platform resources based on a request received from a workload initiator for platform resource candidates, performance capabilities and utilization metrics of the plurality of platform resources, and non-discoverable performance capabilities that are non-discoverable by an operating system, wherein the plurality of platform resources includes one or more graphics processor units (GPUs) and one or more accelerated processing units (APUs); and
dynamically provide a recommendation including the identified one or more platform resource candidates to the workload initiator.
18 . The computer program product of claim 17 , wherein the platform resource candidates are also identified based on the identified one or more platform resource candidates based on non-performance capabilities and performance limitations of the plurality of platform resources.
19 . The computer program product of claim 17 , further comprising computer program instructions that, when executed, cause the computer to:
consult a platform resources library to determine the non-discoverable performance capabilities that are non-discoverable by an operating system, performance limitations, and historical performance data for the plurality of platform resources based on the request from the workload initiator.
20 . The computer program product of claim 17 , further comprising computer program instructions that, when executed, cause the computer to:
dynamically provide, during runtime of the workload, updated platform resource candidates based on adjusted performance capabilities and adjusted utilization metrics of the plurality of platform resources.