Distributed Hashing for Self-Organizing Network Load Balancing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current distributed communication systems with low operating expenditures face inefficiencies in load balancing, node joining times, and data consistency due to redundant links and the 'deer-jump' issue in routing, which affects scalability and performance.

Innovation Solution

A method employing a directed ring topology with a distributed hashing function for identifying storage locations, using reverse redundancy links for data replication, and immediate node integration by transferring data from the predecessor node, allowing for instant joining and reduced routing hops.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If nodes wait passively to be integrated into the network by their predecessor, then data consistency is maintained, but joining time becomes non-deterministic and delays peak request processing

Engineering Contradiction:
Improvedata consistencyVSAvoidjoining time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by allowing nodes to proactively integrate themselves into the network before their predecessor is fully connected. The joining node establishes initial bindings and begins routing requests immediately, rather than waiting passively. This is achieved through the join() method that creates preliminary bindings to predecessor and successor nodes, enabling the node to participate in the network before formal integration is complete.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If redundant links are used for backup purposes, then node failure resilience is improved, but load balancing efficiency deteriorates due to the deer-jump routing issue

Engineering Contradiction:
Improvenode failure resilienceVSAvoidload balancing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies universality by making the successor link serve dual purposes: both as a backup mechanism for node failure and as a routing path for load balancing. Instead of separating backup and routing functions, the successor link is used for both purposes. This eliminates the deer-jump issue where requests would unnecessarily traverse the entire ring, as nodes can now route requests directly through successor links while maintaining backup capabilities.

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

3Reliability

If data is replicated only at successor nodes, then backup capability is ensured, but routing efficiency decreases due to unnecessary hops in the ring topology

Engineering Contradiction:
Improvebackup capabilityVSAvoidrouting efficiency
Core Design Contradiction:
ReliabilityVSSpeed

Solution Approach 1:

The patent applies inversion by changing the direction and purpose of the successor link. Traditionally, successor links were used only for backup in the forward direction, creating inefficiencies. The patent inverts this by allowing nodes to use successor links for routing requests in reverse direction, enabling requests to bypass unnecessary hops through the ring topology and reach destination nodes more directly.

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

Data Source

PatentUS7738466B2Distributed hashing mechanism for self-organizing networks
Publication Date: 2010.06.15 ALCATEL-LUCENT USA LNC
  • US7738466B2 patent drawing
  • US7738466B2 patent drawing
  • US7738466B2 patent drawing

AI summary

A method for retrieving content of a distributed directory of a telecommunication system comprising a plurality of nodes arranged in a directed ring topology, the method comprising the steps of identifying a storage location of said content by means of a distributed hashing function, where the hashing function maps an element onto a node, where the node is responsible carrying elements with a hash value between the node and its successor node. The invention further concerns a corresponding computer software product, telecommunication device, and a telecommunication system.