Flexible RAID Write Cache Policy for Virtual Disk I/O

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current RAID storage systems lack flexibility in write cache policies, forcing users to choose between write-back and write-through modes, which can lead to suboptimal performance for applications requiring a balance between performance and resource conservation.

Innovation Solution

Implementing a flexible write cache policy that allows users to configure virtual disks with a specified proportion of write-back and write-through operations, enabling granular control over IO caching.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If write-back policy is used to improve performance, then IO throughput increases, but cache resources are consumed exclusively by that virtual disk

Engineering Contradiction:
ImproveIO throughputVSAvoidcache resource sharing
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments the write cache policy into two distinct modes (write-back and write-through) that can be independently configured for different virtual disks. This allows the cache resource to be divided and allocated to multiple virtual disks based on their individual performance requirements, rather than being exclusively dedicated to a single disk. Each virtual disk can be configured to use write-back mode for high-performance needs or write-through mode for resource sharing, resolving the contradiction between throughput and resource adaptability.

Inventive Principle:
Principle #1Segmentation

2Reliability

If write-through policy is used to ensure reliability, then data safety is improved, but performance is reduced

Engineering Contradiction:
Improvedata safetyVSAvoidIO throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies local quality by allowing different write cache policies to be configured at the virtual disk level based on individual reliability and performance requirements. Instead of applying a uniform policy across all virtual disks, each disk can be independently configured: critical databases can use write-back for performance, while file servers can use write-through for reliability. This localized configuration resolves the contradiction by matching the policy to the specific operational needs of each virtual disk.

Inventive Principle:
Principle #3Local quality

3Productivity

If all virtual disks use write-back policy, then performance is optimized, but cache bandwidth is exhausted for other disks

Engineering Contradiction:
Improvewrite performanceVSAvoidcache bandwidth consumption
Core Design Contradiction:
ProductivityVSObject-generated harmful factors

Solution Approach 1:

The patent introduces dynamic configuration capability where the write cache policy can be adjusted based on real-time workload conditions and virtual disk priorities. The system can dynamically allocate cache bandwidth to high-priority virtual disks that require write-back performance while ensuring that lower-priority disks can still access the cache through write-through mode. This dynamic adjustment prevents cache bandwidth exhaustion and resolves the contradiction between optimized performance and harmful bandwidth consumption.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12450011B2Flexible write cache policy in RAID environments
Publication Date: 2025.10.21 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD
  • US12450011B2 patent drawing
  • US12450011B2 patent drawing
  • US12450011B2 patent drawing

AI summary

Solutions for providing more customizable performance for a virtual disk in a RAID environment. Some solutions provide for a flexible write cache policy that manages a proportion of input-output operations written in write-back mode against input-output operations written in write-though mode. In some cases, aspects of the flexible write cache policy can be specified by a user.