Log Memory Compression for Bandwidth-Limited Storage Flushes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Storage systems face inefficiencies due to limited bandwidth and CPU utilization in log memory systems, leading to bottlenecks and reduced flush flow parallelism, which limits scalability and performance.

Innovation Solution

A dynamic data compression method that adjusts compression based on storage system bandwidth and CPU utilization, using either the CPU or a separate processing unit to optimize data storage in log memory systems, enhancing flush efficiency and core parallelism.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is compressed for storage in log memory system, then effective bandwidth and log memory size are increased, but CPU utilization increases and processing time is added

Engineering Contradiction:

Improveeffective log memory sizeVSAvoiddata processing time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent applies preliminary compression action by compressing data before it is written to the log memory system. This pre-compression reduces the amount of data that needs to be stored and transferred, effectively increasing the log memory capacity and bandwidth without adding compression overhead during critical path operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary compression layer between the host write requests and the log memory system. This intermediary component handles the compression task, allowing the main storage path to operate more efficiently while the compression function is performed as a separate, optimized operation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If data is compressed using CPU, then compression is performed, but CPU utilization increases causing bottlenecks

Engineering Contradiction:

Improvecompression throughputVSAvoidCPU utilization

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent extracts the compression function from the main CPU processing path by implementing compression as a separate, dedicated function. This extraction allows compression to be performed in parallel or asynchronously, preventing CPU bottlenecks while maintaining compression throughput.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces general-purpose CPU compression with optimized compression mechanisms, potentially using hardware acceleration or specialized compression algorithms that are more efficient and less resource-intensive than standard CPU-based compression.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Quantity of substance

If log memory system bandwidth is limited, then data storage is constrained, but compression increases processing overhead

Engineering Contradiction:

Improvedata storage capacityVSAvoidcompression processing complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent changes the data parameter by transforming it from uncompressed to compressed format, thereby increasing the effective storage capacity of the log memory system. The compression is performed with optimized parameters to minimize processing overhead while maximizing storage efficiency.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12596482B2Log memory compression system and method

Publication Date: 2026.04.07 DELL PROD LP

AI summary

A method, computer program product, and computing system for processing data to store in a storage system. The data is compressed for storage in a log memory system based upon, at least in part, bandwidth of the storage system. The compressed data is stored in a log memory system. The data is stored in a cache memory system in an uncompressed state.