Mobile edge compute dynamic acceleration assignment
Embodiments of a system and method for dynamic hardware acceleration are generally described herein. A method may include identifying a candidate task from a plurality of tasks executing in an operating environment, the operating environment within a hardware enclosure, the candidate task amenable to hardware optimization, instantiating, in response to identifying the candidate task, a hardware component in the operating environment to perform hardware optimization for the task, the hardware component being previously inaccessible to the operating environment, and executing, by the operating environment, a class of tasks amenable to the hardware optimization on the hardware component.
1. A system, in an internet-of-things (IoT) gateway, for dynamic hardware acceleration, the system comprising:
a process monitor to identify a first candidate task and a second candidate task from a plurality of tasks executing in the loT gateway, the first candidate task and the second candidate task being different classes of tasks amenable to a same hardware optimization, the IoT gateway within a hardware enclosure, wherein the IoT gateway is part of a hierarchy of IoT gateways, wherein IoT gateways directly serving IoT devices are placed at a bottom of the hierarchy, IoT gateways serving and served by other IoT gateways are placed in a middle of the hierarchy, and IoT gateways served by a network-edge server are at a top of the hierarchy;
a hardware controller to:
instantiate, in response to identifying the first candidate task, a hardware component to perform hardware optimization for a first class of tasks corresponding to the first candidate task;
fail to instantiate, in response to identifying the second candidate task, the hardware component for a second class of tasks corresponding to the second candidate task because the hardware component is already instantiated for the first class of tasks corresponding to the first candidate task; and
a scheduler, while the hardware component is instantiated for the class of tasks to:
execute tasks corresponding to the first class of tasks;
find a different IoT gateway connected to the IoT gateway in the hierarchy that can instantiate at least one hardware component for the second class of tasks; and
transfer tasks corresponding to the second class of tasks, including the second candidate task, to the different IoT gateway.
2. The system of claim 1 , wherein to identify the first candidate task, the process monitor is to:
measure utilization of a processor; and
identify the first candidate task when the utilization is beyond a threshold.
3. The system of claim 1 , wherein the first candidate task is amenable to hardware optimization when a marker of the first candidate task includes a corresponding class of hardware components in a database.
4. The system of claim 1 , wherein to instantiate the hardware component, the hardware controller is to:
find a component configuration in a database that corresponds to the first class of tasks;
configure the hardware component with the component configuration; and
allocate the configured hardware component to the IoT gateway.
5. A method for dynamic hardware acceleration, the method performed by an internet-of-things (IoT) gateway, the method comprising:
identifying a first candidate task and a second candidate task from a plurality of tasks executing in the IoT gateway, the first candidate task and the second candidate task being different classes of tasks amenable to a same hardware optimization, the IoT gateway within a hardware enclosure, wherein the IoT gateway is part of a hierarchy of IoT gateways, wherein IoT gateways serving IoT devices are placed at a bottom of the hierarchy, IoT gateways serving and served by other IoT gateways are placed in a middle of the hierarchy, and IoT gateways served by a network-edge server are at a top of the hierarchy;
instantiating, in response to identifying the first candidate task, a hardware component to perform hardware optimization for a first class of tasks corresponding to the first candidate;
executing tasks corresponding to the first class of tasks while the hardware component is instantiated for the first class of tasks;
failing, in response to identifying the second candidate task, to instantiate the hardware component for the second class while the hardware component is instantiated for the first class of tasks;
finding a different IoT gateway connected to the IoT gateway in the hierarchy that can instantiate at least one hardware component for the second class of tasks; and
transferring tasks corresponding to the second class of tasks, including the second candidate task, to the different IoT gateway.
6. The method of claim 5 , wherein identifying the candidate task includes:
measuring utilization of a processor; and
identifying the first candidate task when the utilization is beyond a threshold.
7. The method of claim 5 , wherein the first candidate task is amenable to hardware optimization when a marker of the first candidate task includes a corresponding class of hardware components in a database.
8. The method of claim 5 , wherein instantiating the hardware component includes:
finding a component configuration in a database that corresponds to the first class of tasks;
configuring the hardware component with the component configuration; and
allocating the configured hardware component to the IoT gateway.
9. At least one non-transitory machine readable medium including instructions that, when executed, cause an internet-of-things (IoT) gateway to perform operations for dynamic hardware acceleration, the operations comprising:
identifying a first candidate task and a second candidate task from a plurality of tasks executing in the IoT gateway, the first candidate task and the second candidate task being different classes of tasks amenable to a same hardware optimization, the IoT gateway within a hardware enclosure, wherein the IoT gateway is part of a hierarchy of IoT gateways, wherein IoT gateways serving IoT devices are placed at a bottom of the hierarchy, IoT gateways serving and served by other IoT gateways are placed in a middle of the hierarchy, and IoT gateways served by a network-edge server are at a top of the hierarchy;
instantiating, in response to identifying the first candidate task, a hardware component to perform hardware optimization for a first class of tasks corresponding to the first candidate task;
executing tasks corresponding to the first class of tasks while the hardware component is instantiated for the first class of tasks;
failing, in response to identifying the second candidate task, to instantiate the hardware component for the second class while the hardware component is instantiated for the first class of tasks;
finding a different IoT gateway connected to the IoT gateway in the hierarchy that can instantiate at least one hardware component for the second class of tasks; and
transferring tasks corresponding to the second class of tasks, including the second candidate task, to the different IoT gateway.
10. The at least one non-transitory machine readable medium of claim 9 , wherein identifying the candidate task includes:
measuring utilization of a processor; and
identifying the first candidate task when the utilization is beyond a threshold.
11. The at least one non-transitory machine readable medium of claim 9 , wherein the first candidate task is amenable to hardware optimization when a marker of the first candidate task includes a corresponding class of hardware components in a database.
12. The at least one non-transitory machine readable medium of claim 9 , wherein instantiating the hardware component includes:
finding a component configuration in a database that corresponds to the first class of tasks;
configuring the hardware component with the component configuration; and
allocating the configured hardware component to the IoT gateway.