Chunked Data Delivery With Direct-Mapped Flash Addressing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face inefficiencies in managing data storage and retrieval, particularly in flash storage systems, due to redundant write operations and lack of direct control by the operating system, leading to reduced reliability and increased latency.

Innovation Solution

Implementing a direct-mapped flash storage system where the operating system directly addresses data blocks without translation by storage controllers, offloading device management tasks, and utilizing non-volatile RAM for quick data buffering and power failure resilience.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional storage controllers manage data blocks, then device control is centralized, but redundant write operations occur and reliability decreases

Engineering Contradiction:
Improvestorage system reliabilityVSAvoidstorage controller complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the data block management function from the storage controller and transfers it directly to the operating system. The OS now directly addresses data blocks on flash storage without intermediary controller translation, eliminating redundant write operations and improving reliability while reducing controller complexity

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The operating system performs self-service by directly managing data block addresses and write operations. This eliminates the need for controller-mediated translation and coordination, allowing the OS to directly control storage operations and prevent redundant writes that compromised reliability

Inventive Principle:
Principle #25Self-service

2Ease of operation

If storage controllers translate data block addresses, then device management is simplified, but latency increases due to translation overhead

Engineering Contradiction:
Improvedevice management easeVSAvoiddata access latency
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The address translation function is extracted from the storage controller and eliminated entirely. The operating system directly addresses data blocks without translation overhead, achieving both simplified management and reduced latency by removing the translation step that caused time loss

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If flash storage uses traditional write operations, then data can be stored, but redundant writes reduce system performance

Engineering Contradiction:
Improvestorage system performanceVSAvoidenergy consumption
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent implements feedback mechanisms where the operating system tracks existing data block addresses and write operations. This feedback allows the OS to identify and eliminate redundant write operations, improving productivity by preventing unnecessary writes that consume energy and reduce performance

Inventive Principle:
Principle #23Feedback

4Reliability

If non-volatile RAM is used for buffering, then data integrity during power failures is ensured, but additional hardware is required

Engineering Contradiction:
Improvedata integrity during power failureVSAvoidhardware components
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by using non-volatile RAM to buffer data before power failures occur. This pre-positioning of data in non-volatile memory ensures data integrity is maintained automatically during power outages, improving reliability while the integration into existing storage architecture minimizes additional hardware complexity

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12481607B2Optimizing portioned storage data delivery
Publication Date: 2025.11.25 PURE STORAGE INC
  • US12481607B2 patent drawing
  • US12481607B2 patent drawing
  • US12481607B2 patent drawing

AI summary

Chunked data transfers may include: receiving, from an initiator, a read command for first data stored in a storage system; allocating a first buffer of a predefined size; until all of the first data has been sent to the initiator: loading a portion of the first data from the storage system into the first buffer; and sending, to the initiator, the portion of the first data in the first buffer.