Priority Queuing for Flash Storage Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional storage networks relying on rotational media face high access latency due to random data access patterns, which is not efficiently addressed by existing technologies.

Innovation Solution

Implementing priority queuing mechanisms in flash-based storage networks to segregate storage requests by logical units, ensuring high-priority requests are prioritized and processed efficiently, thereby reducing latency and improving overall network performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If priority queuing is implemented in flash-based storage networks, then access latency is reduced and performance is improved, but device complexity increases

Engineering Contradiction:

Improveaccess latencyVSAvoidqueue management complexity

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The storage network is segmented into multiple priority queues (high priority and normal priority) to separate different types of storage requests. This segmentation allows critical requests to be processed independently from standard requests, reducing access latency for time-sensitive operations while maintaining organized queue management.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A priority queueing mechanism acts as an intermediary layer between storage requests and flash-based storage devices. This intermediary sorts and prioritizes requests before they reach the storage medium, reducing latency for high-priority operations without significantly increasing overall system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If traditional rotational media is used in storage networks, then device complexity is reduced, but access latency increases due to random data access patterns

Engineering Contradiction:

Improvestorage device structureVSAvoidaccess latency

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The patent replaces traditional rotational mechanical media with flash-based storage devices that have no moving parts. This substitution eliminates the fundamental limitation of rotational latency and mechanical seek times, dramatically reducing access latency for random data access patterns while maintaining acceptable device complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The storage medium parameter is changed from rotational magnetic media to solid-state flash memory. This parameter change fundamentally alters the access characteristics, enabling much faster random access times and eliminating the trade-off between device simplicity and access speed that plagues traditional rotational systems.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS9983801B1Priority queueing for low latency storage networks

Publication Date: 2018.05.29 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD

AI summary

A method for a storage area network includes setting a priority indicator in a signal of a storage request to form a high priority storage request for a storage target in the storage area network. The method further includes transmitting the high priority storage request to the storage target before the transmission of at least one normal priority storage request. The high priority storage request then traverses the storage area network with a higher priority than a normal priority storage request to the storage target. Upon receiving the high priority storage request, the storage target executes the high priority storage request prior to executing at least one normal priority storage request.