Multi-tier Distributed System Transaction Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional multi-tier distributed systems face challenges in enhancing the independence of tiers, requiring large storage for long transactions and complicating data management due to the need for inter-tier processing, which hinders efficient operation.

Innovation Solution

A multi-tier distributed system is designed with monitoring and communication mechanisms between upper and lower nodes, allowing for independent execution of transactions across tiers, including version management and dynamic node updating, to enhance tier independence and simplify data management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If processing of one tier is inserted into processing of another tier to guarantee atomicity, then transaction atomicity is improved, but independence of each tier deteriorates

Engineering Contradiction:
Improvetransaction atomicityVSAvoidindependence of each tier
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent segments the distributed transaction processing into independent tier-level transactions, where each tier processes its own transaction independently. The upper tier executes its transaction and the lower tier executes its transaction separately, with each having its own transaction management. This segmentation allows each tier to maintain independence while still achieving overall atomicity through the coordination mechanism.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If one transaction takes a long time, then storage capacity required is improved, but management and maintenance difficulty increases

Engineering Contradiction:
Improvestorage capacityVSAvoidmanagement and maintenance
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent divides a long-running distributed transaction into multiple shorter tier-level transactions that execute independently at different tiers. Instead of holding all resources in memory for the entire duration of a long transaction, each tier processes and commits its transaction separately, reducing the storage burden and improving manageability.

Inventive Principle:
Principle #1Segmentation

3Reliability

If inter-tier processing is required to ensure transaction consistency, then transaction consistency is improved, but system complexity increases

Engineering Contradiction:
Improvetransaction consistencyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where the upper node monitors the running status of the lower node through monitoring means. The lower node includes monitor enabling means that allow the upper node to monitor its status. Based on this feedback, the system can dynamically adjust transaction execution, ensuring consistency while managing complexity through intelligent coordination rather than rigid inter-tier processing.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS9015308B2Multilayer distributed processing system
Publication Date: 2015.04.21 VALTRUS INNOVATIONS LTD
  • US9015308B2 patent drawing
  • US9015308B2 patent drawing
  • US9015308B2 patent drawing

AI summary

The independencies of a plurality of layers executing dividingly a transaction can be easily enhanced. Anode (30) assigns to a transaction to anode (30) of a lower layer through a distributed transaction management section. The node (30) shares a predetermined transaction with the node (30) of the lower layer along with other nodes (30). The node (30) shared by the nodes (30) is a read-only node or a node to which data can be written by the characteristic of a function. Thus the node (30) searches for an unused node (30) in lower layers through the distributed transaction management section (34) when the node starts a new transaction. First, second, and third node hosts (3,4,5) check if each node (30) is used for which transaction or if each node (30) is used or not and store the results.