Storage Controller Power Prediction and Dynamic Resource Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage devices, such as disk arrays, face challenges in optimizing power consumption while maintaining performance, as existing systems lack efficient methods to predict and adjust power usage based on varying access patterns and throughput demands.
Innovation Solution
A memory system that includes a controller and a nonvolatile semiconductor memory, which predicts power consumption based on access patterns and throughput from a host, negotiates power resources, and adjusts performance levels by controlling parallel operating resources to operate within permissible power limits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the storage device operates at high performance levels to meet throughput demands, then productivity is improved, but power consumption increases
Solution Approach 1:
The storage device dynamically adjusts its operating performance level based on predicted power consumption and permissible power constraints. The controller monitors access patterns, calculates required power consumption, and adapts the performance level accordingly, allowing the system to operate at high performance when power is available and reduce performance when power constraints are detected, thus resolving the contradiction between maintaining high productivity and managing power consumption
Solution Approach 2:
The system changes operational parameters by adjusting the performance level of the storage device based on predicted power consumption. By modifying parameters such as cache size, buffer depth, and parallel operation degrees, the system can achieve high throughput when permissible power is sufficient while consuming less power when constraints are imposed, effectively balancing productivity and power usage
2Productivity
If the storage device increases parallel operating resources to improve performance, then productivity is improved, but device complexity increases
Solution Approach 1:
The controller dynamically adjusts the degree of parallel operations based on the target performance level and predicted power consumption. Instead of maintaining fixed high-level parallel resources that would increase device complexity, the system activates only the necessary number of parallel operations required to achieve the current performance target, thereby improving productivity when needed while avoiding unnecessary complexity
3Use of energy by moving object
If the storage device uses limited power resources to maintain operation, then power consumption is reduced, but productivity deteriorates
Solution Approach 1:
When power consumption needs to be reduced, the system changes operational parameters by lowering the performance level and reducing the degree of parallel operations. The controller adjusts cache sizes, buffer depths, and other operational parameters to achieve lower power consumption while maintaining acceptable functionality, thus resolving the contradiction between power conservation and productivity
Data Source
AI summary
A storage device includes a nonvolatile semiconductor memory and a controller. The controller is configured to predict power consumption that is required to carry out operations in accordance with access pattern and throughput received from a host, notify the predicted power consumption to the host, determine operating resources of at least one of the nonvolatile semiconductor memory and the controller to carry out the operations, on the basis of the permissible power consumption received from the host, and carry out the operations using the determined operating resources.