DHT Load Balance via Virtual Identifier Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In Distributed Hash Table (DHT) networks, managing and transferring virtual identifiers becomes cumbersome due to each node being responsible for a large number of identifiers, leading to workload inefficiencies and load imbalance.

Innovation Solution

A method to determine and allocate a specific number of virtual identifiers to each node based on its ability, ensuring load balance by equally dividing the load space and adjusting identifiers when new nodes are added or when load imbalance exceeds a threshold, using a formula to calculate optimal virtual identifier distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the whole Hash space is equally divided into Q portions with Q virtual identifiers (where Q is much greater than the quantity S of nodes), then load balance is achieved, but the workload for managing and transferring virtual identifiers increases significantly

Engineering Contradiction:
Improvenumber of virtual identifiersVSAvoidworkload efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent segments the virtual identifier management by introducing a two-level hierarchy: global virtual identifiers (GVIDs) for overall load balance and local virtual identifiers (LVIDs) for node-specific management. This segmentation reduces the number of identifiers each node must manage while maintaining load balance across the network.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a coordinator node as an intermediary that manages the global virtual identifier space and allocates portions to individual nodes. This intermediary handles the complexity of global identifier management, freeing individual nodes from managing large numbers of identifiers while maintaining overall load balance.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If each node is allocated many virtual identifiers to cover the Hash space, then load balance is improved, but the complexity of managing and transferring these identifiers increases

Engineering Contradiction:
Improvenumber of virtual identifiers per nodeVSAvoidmanagement complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent divides the virtual identifier space into global identifiers managed by a coordinator and local identifiers managed by individual nodes. Each node only needs to manage a small subset of LVIDs locally, while GVIDs are handled by the coordinator, significantly reducing per-node management complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The coordinator node serves multiple functions: managing the global virtual identifier space, allocating identifiers to nodes, and handling load balance adjustments. This multi-functionality consolidates complex management tasks in a single entity, reducing the burden on individual nodes.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Quantity of substance

If virtual identifiers are allocated according to node ability, then load balance is achieved, but the system becomes less adaptable when nodes are added or removed

Engineering Contradiction:
Improvedistribution of virtual identifiersVSAvoidnetwork scalability
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic virtual identifier allocation where the coordinator continuously monitors node status and reallocates GVIDs as nodes are added or removed. The system adapts to changing network conditions by dynamically adjusting the distribution of virtual identifiers based on current node capacity and load, rather than using static allocation.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where nodes report their load status to the coordinator, which then adjusts virtual identifier allocation accordingly. This feedback loop enables the system to automatically adapt to changes in network topology and node capacity, maintaining load balance while supporting scalability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2632111B1Method and apparatus for realizing load balance in distributed hash table network
Publication Date: 2019.04.03 ZTE CORP
  • EP2632111B1 patent drawingFigure 1
  • EP2632111B1 patent drawingFigure 2~3
  • EP2632111B1 patent drawingFigure 4~5

AI summary

The present invention discloses a method and an apparatus for realizing load balance in a DHT network, wherein the method comprises: determining the quantity Z of first virtual identifiers according to the node planning of the DHT network and a preset threshold of load unbalance degree, wherein Z is a natural number; equally dividing the whole load space of the DHT network into Z portions, and each portion of the load space is respectively identified by a different first virtual identifier; allocating the Z first virtual identifiers to each node according to the ability of each node so that each node is in charge of the load space corresponding to the first virtual identifier allocated to the node. The present invention can reduce workload for managing and transferring virtual identifiers while realizing data storage load balance in the DHT network.