Partitioned Processor Power Management With Dedicated DMA Engines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing power consumption of modern integrated circuits leads to higher system costs due to the need for costly cooling systems, which is a challenge across various computing devices, including portable computers, mobile communication devices, desktop computers, and servers, especially with the complexity of powering down systems with multiple sockets.
Innovation Solution
A processing unit is divided into two partitions, each with dedicated direct memory access (DMA) engines and operational parameters, allowing tasks to be assigned based on type, with one partition handling real-time data processing and the other non-real-time data processing, thereby optimizing power consumption by adjusting operational parameters and disconnecting idle partitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple processor cores and pipelines are integrated to increase performance, then processing capability is improved, but power consumption increases
Solution Approach 1:
The processing unit is divided into multiple partitions, each with dedicated DMA engines. This segmentation allows independent power management of each partition, enabling the system to activate only the necessary partitions for current tasks, thus reducing overall power consumption while maintaining processing capability.
Solution Approach 2:
The system dynamically adjusts operational parameters and power states of different partitions based on task requirements. By transitioning partitions between active and idle states with different power levels, the system optimizes the balance between processing capability and power consumption in real-time.
2Productivity
If multiple sockets with copies of selected packages are used to scale system performance, then processing power is improved, but power management complexity increases
Solution Approach 1:
Each socket or processing package is divided into multiple partitions with dedicated DMA engines, allowing fine-grained power management within each socket. This reduces the complexity of managing entire sockets by enabling selective powering down of unused partitions while keeping others active.
Solution Approach 2:
Each partition is equipped with dedicated DMA engines that can independently manage data transfer and power states. This self-service capability reduces the coordination overhead required for power management across multiple sockets, as each partition can autonomously transition to low-power states when idle.
3Productivity
If dedicated DMA engines are assigned to each partition, then data transfer efficiency is improved, but device complexity increases
Solution Approach 1:
Dedicated DMA engines are assigned to specific partitions to enable independent data transfer operations. This segmentation improves data transfer efficiency by eliminating contention between partitions while the modular architecture keeps each DMA engine relatively simple, managing the complexity through systematic distribution.
Solution Approach 2:
Each dedicated DMA engine is designed with universal functionality to handle various data transfer tasks within its partition. This multi-functionality reduces the need for specialized hardware for each transfer type, thereby limiting the increase in device complexity while maintaining high transfer efficiency.
Data Source
AI summary
Apparatuses, systems and methods for performing efficient power management for a processing unit. A processing unit includes two partitions, each assigned to a respective power domain with operating parameters, and each with a respective direct memory access (DMA) engine. If a controller determines a task type of a received task indicates the task is to be processed by components of the second partition, then the controller assigns the task to the second partition and maintains the operational parameters of the first power domain for the components of the first partition or selects lower performance operational parameters of the first power domain. The processing unit accesses data stored in memory using a DMA engine and operational parameters of the second partition. Additionally, the second partition processes the task using the operational parameters of the second power domain.