Multi-Plane SSD Read Scheduling for Lower Power Use
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Solid-state drives (SSDs) with multi-plane non-volatile memory (MPNVM) devices consume excessive power during read operations due to the operation of both general circuitry and specific planes, which is particularly problematic for battery-powered devices.
Innovation Solution
Implement mechanisms to execute multiple read commands on different planes of the MPNVM SSD simultaneously, reducing peripheral and plane array power consumption by overlapping command execution times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If multiple read commands are executed sequentially on different planes, then power consumption is reduced by operating only one plane at a time, but the total execution time increases
Solution Approach 1:
The system dynamically adjusts command execution strategy based on plane availability and command priority. It transitions from static sequential execution to dynamic parallel execution when multiple planes are available, optimizing the trade-off between power consumption and execution time in real-time
Solution Approach 2:
The system implements periodic checking of plane availability and command queues, enabling it to periodically switch between parallel and sequential execution modes based on current system state, thus balancing power efficiency with performance requirements
2Productivity
If multiple read commands are executed in parallel on different planes, then command execution time is reduced, but power consumption increases due to simultaneous operation of multiple planes
Solution Approach 1:
The system applies partial parallelism by executing only as many commands in parallel as necessary to meet performance targets, rather than maximizing parallel execution. It selectively parallelizes commands based on priority and timing requirements, avoiding unnecessary power consumption from excessive parallel operation
Solution Approach 2:
The system changes operational parameters dynamically by adjusting the degree of parallelism based on command priority, plane availability, and power state. It modifies execution parameters to optimize the balance between throughput and power consumption for different workloads
3Use of energy by moving object
If the SSD waits for commands to be batched together, then power consumption is reduced by consolidating operations, but command latency increases
Solution Approach 1:
The system performs preliminary assessment of command characteristics and plane availability before executing commands. It proactively identifies opportunities for power-efficient batching while maintaining speed requirements through advance planning and command prioritization
Solution Approach 2:
The system implements feedback mechanisms that monitor command completion status, plane availability, and power consumption metrics. This feedback enables dynamic adjustment of batching strategies to maintain acceptable latency while optimizing power efficiency
Data Source
AI summary
Mechanisms (which can include systems, methods, and media) for reducing power consumption of multi-plane non-volatile-memory (MPNVM) solid-states drives (SSDs) are provided. The mechanisms include: identifying a first MPNVM device and a first plane of the first MPNVM device associated with a first command; determining that one or more other commands are available for the first MPNVM device and for a plane of the first MPNVM device different from the first plane; and in response to determining that the one or more other commands are available for the first MPNVM device and for a plane of the first MPNVM device different from the first plane, sending the first command and the one or more other commands to the first MPNVM device for execution.