Storage Device Memory Scheduling for Power Budget Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-capacity and high-integration storage devices has highlighted the need for effective power management to reduce power consumption.
Innovation Solution
A storage device with a storage controller that determines the operation order of memory devices based on a power consumption profile table, considering real-time power consumption and power budget, to optimize power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If high-capacity and high-integration memory devices are used, then storage capacity is improved, but power consumption increases
Solution Approach 1:
The memory system is divided into multiple independent memory devices, each with its own power consumption characteristics. The controller segments the command queue and processes commands in batches, allowing selective activation of memory devices based on power budget availability. This segmentation enables the system to maintain high storage capacity while managing power consumption at the device level.
Solution Approach 2:
The power management system dynamically adjusts the operation order and batch size of memory device commands based on real-time power consumption profiles and current power budget status. The controller continuously monitors power usage and adapts its command scheduling strategy, transitioning between different operational modes to optimize the balance between storage utilization and power consumption.
2Loss of energy
If power management based on die characteristics is implemented, then power usage efficiency is improved, but system complexity increases
Solution Approach 1:
Power consumption profiles for each memory device are pre-characterized and stored in the controller before actual operation begins. These profiles capture the specific power consumption characteristics of each die, allowing the controller to make informed scheduling decisions without performing complex real-time measurements. This preliminary characterization simplifies the runtime power management logic.
Solution Approach 2:
The system implements a feedback mechanism where the controller monitors actual power consumption against predicted values from the profiles, and adjusts the command scheduling strategy accordingly. This closed-loop control enables the system to maintain optimal power efficiency while adapting to varying operational conditions, without requiring overly complex predictive algorithms.
3Productivity
If multiple memory devices are operated simultaneously, then productivity is improved, but power consumption exceeds budget
Solution Approach 1:
Instead of operating all memory devices simultaneously or sequentially processing commands one at a time, the system processes commands in optimized batches that partially activate the memory device array. The controller determines the optimal batch size and composition based on power consumption profiles and current power budget status, achieving a compromise between throughput and power consumption by operating a subset of devices at any given time.
Solution Approach 2:
The power management system employs periodic batching of command execution, where memory devices are activated in regular intervals rather than continuously. This periodic operation allows the system to accumulate power budget during low-activity periods and utilize it during command execution batches, effectively decoupling instantaneous power demand from average power consumption and enabling higher overall productivity within power constraints.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A storage device 101 includes a plurality of memory devices 120 and a storage controller 110 receiving a plurality of commands that operate the plurality of memory devices 120 from an external host device. The storage device 101 determines an operation order of the plurality of memory devices 120 according to the plurality of commands based on a power consumption profile table including a power information indicating a power consumption of each of the plurality of memory devices 120.