Metadata Tree Node Splitting for Lock Contention Relief

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Lock contention in tree data structures of storage systems leads to delays and increased processing load, particularly in frequently accessed nodes, which existing self-balancing mechanisms fail to adequately address.

Innovation Solution

Implement a contention-based split mechanism that splits hot nodes into sibling nodes based on access frequency and lock contention, even if the node is not yet full, and tags these nodes to avoid merging until a defined period, balancing access load and reducing lock contention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a node is split only when full (traditional self-balancing mechanism), then the tree structure remains balanced, but lock contention increases on frequently accessed nodes

Engineering Contradiction:

Improvetree balanceVSAvoidlock contention

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent divides a frequently accessed node (hot node) into multiple child nodes based on access frequency patterns, even when the node is not full. This segmentation distributes the access load across multiple nodes, reducing lock contention on any single node while maintaining the tree structure's integrity

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces dynamic node splitting based on runtime access patterns rather than static fullness thresholds. The system monitors access frequency and dynamically creates child nodes to adapt to changing workloads, allowing the tree structure to evolve with usage patterns

Inventive Principle:

Principle #15Dynamics

2Productivity

If hot nodes are split into multiple child nodes, then lock contention is reduced, but processing load and complexity increase

Engineering Contradiction:

Improvelock contentionVSAvoidprocessing load

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary analysis of access patterns to identify hot nodes before splitting them. By pre-processing access statistics and identifying candidates for splitting in advance, the system reduces the complexity of real-time decisions during node access operations

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism that monitors access patterns and lock contention metrics to guide the node splitting process. The system uses this feedback to continuously optimize the tree structure, splitting nodes when beneficial and potentially merging them when access patterns change

Inventive Principle:

Principle #23Feedback

3Productivity

If nodes are split based on access frequency, then system performance improves, but the tree structure becomes more complex to manage

Engineering Contradiction:

Improvesystem performanceVSAvoidtree structure management

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent applies different splitting strategies to different parts of the tree based on local access patterns. Each node is evaluated independently, and splitting decisions are made locally based on that node's specific access characteristics rather than applying a global rule to the entire tree

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12566745B1Splitting a node of a tree data structure

Publication Date: 2026.03.03 VAST DATA LTD

AI summary

A method for splitting a node of a metadata tree, the method includes (i) splitting the node of the metadata tree based on a lock contention parameter that differs from a fullness of the node, to provide split nodes; and (ii) preventing a size based merge of the split nodes for a time period following a period of a defined duration that starts after the splitting.