Memory Controller Token Management for Read QoS

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Memory systems face challenges in maintaining read quality of service (QoS) while performing background operations, as existing memory controllers do not effectively manage the prioritization and execution of foreground and background operations to ensure high read performance and system management tasks are balanced.

Innovation Solution

A memory controller that counts read operations and generates tokens to allow background operations, prioritizing them based on a token management system to ensure that background operations are performed without compromising read QoS, by allocating tokens to perform background tasks only when necessary and ensuring they do not interfere with read operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If background operations are performed frequently to maintain memory system management, then system reliability is improved, but read QoS deteriorates due to resource contention

Engineering Contradiction:

Improvesystem management reliabilityVSAvoidread performance quality

Core Design Contradiction:

ReliabilityVSManufacturing precision

Solution Approach 1:

The patent implements a token generation mechanism that proactively creates tokens based on predicted read operation patterns. The controller generates tokens in advance during periods of low read activity, so they are readily available when background operations need to be performed, preventing read QoS degradation while maintaining system management reliability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adjusts background operation execution based on real-time token availability and read operation intensity. The controller monitors read operation patterns and adaptively determines when to execute background operations, transforming the static priority scheme into a dynamic resource allocation mechanism that responds to changing system conditions

Inventive Principle:

Principle #15Dynamics

2Productivity

If background operations are prioritized to ensure memory management, then productivity is improved, but read operation speed deteriorates

Engineering Contradiction:

Improvebackground operation throughputVSAvoidread operation speed

Core Design Contradiction:

ProductivityVSSpeed

Solution Approach 1:

The patent introduces tokens as an intermediary resource that mediates between background operations and read operations. Tokens act as permission certificates that allow background operations to execute without directly competing for memory bandwidth with read operations, enabling productivity improvement while protecting read operation speed through indirect resource allocation

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the control parameter from direct priority assignment to token-based permission control. By transforming the resource allocation mechanism from a static priority queue to a dynamic token consumption model, the system can increase background operation throughput while maintaining read operation speed through parameter-driven resource management

Inventive Principle:

Principle #35Parameter changes

3Manufacturing precision

If read operations are strictly prioritized to maintain read QoS, then read performance is improved, but background operation execution is delayed

Engineering Contradiction:

Improveread QoS complianceVSAvoidbackground operation latency

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent implements periodic token generation based on read operation intensity patterns. During low-intensity read periods, tokens are generated and accumulated, enabling background operations to execute without interfering with read QoS. This periodic resource allocation reduces background operation latency while maintaining strict read QoS compliance through rhythm-based resource management

Inventive Principle:

Principle #19Periodic action

4Adaptability or versatility

If dynamic resource allocation is implemented to balance foreground and background operations, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improveoperation prioritization flexibilityVSAvoidcontroller structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a self-service token management system where the controller automatically generates, distributes, and tracks tokens without external intervention. The token manager autonomously monitors read operation patterns, generates tokens accordingly, and controls background operation execution, providing adaptability while minimizing complexity through self-managing architecture

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10885995B2Memory controller, memory system including memory controller, method of operating memory controller

Publication Date: 2021.01.05 SK HYNIX INC

AI summary

A memory controller for use in a memory system may include a counter configured to count a number of times a read operation corresponding to a read request received from a host is performed; a token manager configured to generate a token each time a count value of the counter reaches a preset count, the token representing a right to perform a background operation; and a operation performing unit configured to perform foreground operations corresponding to the respective requests in response to the requests received from the host, request the token manager to allocate the token to the operation performing unit each time the background operation is triggered, and perform the background operation when the token is allocated from the token manager to the operation performing unit.