Abstract:
A computing device includes a multi-core processor which includes two or more cores, each having a significant task list queue and a normal task list queue, and a scheduler which allocates a task to each of the two or more cores. The scheduler, following allocation of the tasks, receives a new task, selects one core among the two or more cores based on task information of the new task, and allocates the new task to the selected core, when the allocation of the new task to the selected core will not change an execution schedule of a significant task in the significant task list queue of the selected core.
Abstract:
A semiconductor device is provided. The semiconductor device includes a processing device that provides resource usage information including a utilization value; and a prediction information generating device that generates resource usage prediction information based on the resource usage information and provides the resource usage prediction information to the processing device. The prediction information generating device includes: an error calculator to calculate an error value between the utilization value and a predicted value included in the resource usage prediction information; a margin value calculator to receive the error value from the error calculator and calculate a margin value using the error value; an anchor value calculator to calculate an anchor value using the utilization value; and a predictor to output the predicted value using the anchor value and the margin value. The processing device controls resource allocation of the processing device based on the resource usage prediction information.
Abstract:
A scheduling method includes calculating required performance for a given task, calculating use performance and real performance of a candidate processor, calculating power corresponding to the real performance, calculating expected energy usage of the candidate processor based on the required performance, the use performance, the real performance, and the calculated power and calculating performance efficiency of the candidate processor by considering a ratio of the expected energy usage to the real performance.
Abstract:
A computing device and a method for operating the computing device are provided. The computing device includes a task classifying module and a task allocating and managing module. The task classifying module classifies a task scheduled to be processed by a multi-core processor into an expectable task or a normal task. The task allocating and managing module selects one core of the multi-core processor as a target core and allocates the task scheduled to be processed to the target core. In response to the task scheduled to be processed being classified as the normal task, the task allocating and managing module determines whether to allocate the normal task to the target core depending on whether the target core processes a previously allocated expectable task.
Abstract:
A power management device includes a workload rate detector configured to adjust a length of a duration period; a power management unit configured to calculate a period workload rate in the duration period; and a voltage-clock provider configured to adjust a power level, based on the period workload rate and/or based on an external command.
Abstract:
A method of operating a system-on-chip (SOC) including a central processing unit (CPU) and a target hardware to which a dynamic voltage and frequency scaling (DVFS) is applied, includes determining an operating scheme of the target hardware, setting a DVFS application scheme for applying the DVFS to the target hardware, based on the operating scheme of the target hardware, and performing the DVFS on the target hardware, based on the DVFS application scheme.
Abstract:
A power management device includes a workload rate detector configured to adjust a length of a duration period; a power management unit configured to calculate a period workload rate in the duration period; and a voltage-clock provider configured to adjust a power level, based on the period workload rate and/or based on an external command.
Abstract:
A semiconductor device is provided. The semiconductor device includes a processing device that provides resource usage information including a utilization value; and a prediction information generating device that generates resource usage prediction information based on the resource usage information and provides the resource usage prediction information to the processing device. The prediction information generating device includes: an error calculator to calculate an error value between the utilization value and a predicted value included in the resource usage prediction information; a margin value calculator to receive the error value from the error calculator and calculate a margin value using the error value; an anchor value calculator to calculate an anchor value using the utilization value; and a predictor to output the predicted value using the anchor value and the margin value. The processing device controls resource allocation of the processing device based on the resource usage prediction information.
Abstract:
Provided are multi-core control systems. A multi-core control system includes multiple cores including a first core; and a process dependency recognizer configured to recognize a dependency between processes each executed in the respective cores, wherein if the first core waits for a first period of time to execute a first process, the first core recognizes a process on which the first process depends by the process dependency recognizer.
Abstract:
A method of dynamically scaling a power level of a microprocessor is provided. The method includes: receiving a plurality of workload rates of a microprocessor in a first duration period; determining a second duration period by adjusting a length of the first duration period; calculating a period workload rate based on the plurality of workload rates in the first duration period; dynamically scaling a power level of the microprocessor; and maintaining the scaled power level of the microprocessor in the second duration period.