Flash Translation Layer Power Allocation for Data Storage

Solution Overview

Problem

Data storage devices face challenges in managing power distribution efficiently, particularly when the current demand exceeds the available power supply, leading to voltage drops that can render some memory devices non-functional, especially with the use of non-prime NAND dies and 4-Bit-Per-Cell memories which are prone to errors and early failure.

Innovation Solution

A flash translation layer (FTL) dynamically prioritizes memory operations to allocate power more effectively, ensuring that higher-priority operations, such as read-retry and read-scrub-based relocation, receive preferential power allocation, while maintaining optimal data throughput and balancing input/output bus utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If parallel processing is used to increase performance, then data throughput improves, but power consumption increases and exceeds the power budget

Engineering Contradiction:

Improvedata throughputVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The system dynamically adjusts the number of active storage dies based on real-time power availability and operational priority. The controller monitors power budget constraints and dynamically powers up or down storage dies, changing the parallel processing capacity from fixed to adaptive. This resolves the contradiction by allowing the system to maintain high throughput when power is available while reducing consumption when constrained.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Different storage dies are assigned different operational characteristics and power requirements. Critical operations are prioritized on specific dies while less critical operations are deferred. The system applies local quality differentiation by treating different storage resources differently based on their operational importance, allowing optimized power distribution that maintains performance for essential functions while conserving power overall.

Inventive Principle:

Principle #3Local quality

2Productivity

If more storage dies are powered up to increase throughput, then data processing capability improves, but voltage drops occur when current demand exceeds supply

Engineering Contradiction:

Improvedata processing capabilityVSAvoidvoltage stability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts the number of active storage dies based on real-time power availability and operational priority. The controller monitors power budget constraints and dynamically powers up or down storage dies, changing the parallel processing capacity from fixed to adaptive. This resolves the contradiction by allowing the system to maintain high throughput when power is available while reducing consumption when constrained.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The controller performs preliminary assessment of power availability and operational priorities before allocating power to storage dies. By evaluating the power budget and operation priorities in advance, the system pre-determines which storage dies should be powered up, preventing voltage drops before they occur and ensuring stable operation.

Inventive Principle:

Principle #10Preliminary action

3Device complexity

If default scheduling scheme is used, then implementation simplicity is maintained, but performance optimization is limited

Engineering Contradiction:

Improvescheduling implementationVSAvoideffective data throughput

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

Different storage dies are assigned different operational characteristics and power requirements. Critical operations are prioritized on specific dies while less critical operations are deferred. The system applies local quality differentiation by treating different storage resources differently based on their operational importance, allowing optimized power distribution that maintains performance for essential functions while conserving power overall.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The controller continuously monitors power consumption, voltage levels, and operational status to adjust the scheduling decisions in real-time. This feedback mechanism allows the system to learn from actual performance data and refine its scheduling algorithm, improving throughput optimization while maintaining manageable implementation complexity through iterative improvement.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11941263B2Flash-translation-layer-aided power allocation in a data storage device

Publication Date: 2024.03.26 SANDISK TECHNOLOGIES LLC

AI summary

A data storage device having an FTL configured to award to some pending memory operations a higher priority compared to the priority given to those operations by a default scheduling scheme. Such awards of higher priority may be based on a policy directed, e.g., at maximizing the effective data throughput, balancing the data throughput and the input/output bus throughput, or other performance objective. In response to awards of higher priority, a power-management circuit of the data storage device may dynamically route a constrained power supply such that the storage dies corresponding to the higher-priority operations preferentially receive power allocation in the next time interval(s). The remainder of the power budget (if any) in those time intervals may be allocated in accordance with the default scheduling scheme. According to an embodiment, nonlimiting examples of higher-priority operations may include read-retry operations, read-scrub-based relocation, internal control operations, or other suitable higher-priority operations.