Storage Volume Discard Control for Low-Latency Data Deletion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems face inefficiencies in data deletion processes, particularly in flash storage systems, where lower-level storage controllers perform redundant operations, leading to increased latency and reduced reliability.

Innovation Solution

Implementing a direct-mapped flash storage system where higher-level operating systems initiate and control data deletion processes directly, eliminating the need for lower-level storage controller involvement, thereby optimizing data deletion operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If lower-level storage controllers perform data deletion operations, then data deletion can be executed, but redundant operations occur leading to increased latency and reduced reliability

Engineering Contradiction:
Improvedata deletion reliabilityVSAvoiddata deletion latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts the data deletion control function from lower-level storage controllers and relocates it to higher-level operating systems. The storage controller is reduced to performing only execution of deletion commands received from the OS, eliminating redundant operations at the controller level and reducing deletion latency while improving reliability through centralized control.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent inverts the traditional control hierarchy by having the operating system (higher level) initiate and control data deletion operations instead of the storage controller (lower level). This inversion eliminates the redundancy where both levels attempted to manage deletion, thereby reducing latency and improving reliability through a single point of control.

Inventive Principle:
Principle #13The other way round (Inversion)

2Productivity

If multiple storage controllers are involved in data deletion, then data can be deleted across multiple flash drives, but unnecessary write operations increase reducing efficiency

Engineering Contradiction:
Improvedata deletion efficiencyVSAvoidwrite operations overhead
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent extracts the decision-making logic for data deletion from storage controllers and centralizes it in the operating system. The OS determines which data blocks need deletion and sends targeted deletion commands to storage controllers, eliminating unnecessary write operations that occurred when controllers independently managed deletion across multiple flash drives.

Inventive Principle:
Principle #2Taking out (Extraction)

3Ease of operation

If lower-level storage controllers control data deletion, then deletion operations can be performed, but system complexity increases due to coordination requirements

Engineering Contradiction:
Improvedata deletion operation simplicityVSAvoidstorage controller coordination complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent extracts the coordination complexity of data deletion operations from the storage controller layer and moves it to the operating system layer. The OS handles the complex logic of determining which data to delete and coordinates with storage controllers only for execution, simplifying the storage controller's role and reducing system complexity at the hardware level.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS12578858B2Load-based data deletion optimization in a storage system
Publication Date: 2026.03.17 PURE STORAGE INC
  • US12578858B2 patent drawing
  • US12578858B2 patent drawing
  • US12578858B2 patent drawing

AI summary

An example method for optimizing data deletion in a storage system comprises: monitoring a load metric associated with a storage volume using a file system, the load metric indicative of a dynamically changing operational load of the storage volume; determining, based on the monitored load metric, a discard rate; and initiating, based on detecting a command to delete data from the storage volume and based on the discard rate, a discard request for the storage device, the discard request commanding the storage device to indicate that one or more blocks storing the data within a storage device associated with the storage volume are free for use by the file system to store additional data or available to be reclaimed by a garbage collection process.