Memory Centric Architecture Raw Data Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data storage systems face challenges in efficiently adapting to application usage without relying on hinting, tracing, or user input, leading to lagged information and inefficient data storage management.

Innovation Solution

Implementing a Memory Centric Architecture (MCA) that enables unilateral classification of data within the memory footprint from raw bits, allowing for the discovery of application usage without hints or user input, using techniques such as sampling and data science methods to identify and categorize data, thereby improving data storage system performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data storage systems rely on hinting, tracing, or user input to adapt to application usage, then the system can make informed data placement decisions, but the system complexity increases and real-time adaptation is lost due to lagged information

Engineering Contradiction:

Improveadaptability to application usageVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The storage system performs unilateral classification of data within the memory footprint from raw bits without requiring hints, tracing, or user input. The system independently discovers application usage patterns by analyzing data characteristics directly, enabling self-service adaptation that reduces complexity while maintaining real-time responsiveness

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary classification and identification of data objects from raw data before data placement decisions are made. By pre-analyzing data characteristics and categorizing objects in advance, the system prepares information needed for optimal data placement without waiting for runtime hints or user input, reducing adaptation lag

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If the system performs unilateral classification from raw bits without hints or user input, then real-time identification is achieved, but the difficulty of detecting and measuring data characteristics increases

Engineering Contradiction:

Improvereal-time identification capabilityVSAvoiddifficulty of detecting data characteristics

Core Design Contradiction:

Loss of timeVSDifficulty of detecting and measuring

Solution Approach 1:

The system applies sampling techniques to analyze portions of raw data rather than processing entire data sets. By examining representative samples of data characteristics, the system achieves real-time identification without the computational burden of complete analysis, making detection and measurement feasible while maintaining temporal responsiveness

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system replaces traditional mechanical approaches of hinting and tracing with data science methods that automatically detect and measure data characteristics from raw bits. By using computational algorithms for pattern recognition and classification, the system simplifies the detection process while achieving real-time performance

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10394472B1Classification and identification from raw data within a memory domain

Publication Date: 2019.08.27 EMC IP HLDG CO LLC

AI summary

A system, computer program product, and computer-executable method of managing one or more tiers of memory of a host computing system, the system, computer program product, and computer-executable method including accessing a portion of raw data from a memory page associated with data stored on the one or more tiers of memory, sampling the portion of raw data to select a sample data, analyzing the sample data to determine a sample category, and classifying the portion of raw data based at least in part by considering the sample category.