Load Balancer Service Request Processing with Data Node Status Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In telecommunications service processing systems, when a fault occurs in a data node, it leads to session queue congestion and reduced efficiency in processing service requests, as the application node cannot handle requests from other normal data nodes in a timely manner due to occupied session queues.

Innovation Solution

A service request processing method that involves an application node receiving a request, determining the status of a data node by calculating delays and comparing them to a fault threshold, and sending evaluation responses to an evaluation center or load balancer to manage the status of data nodes, thereby preventing congestion by rejecting or temporarily storing requests when a data node is unavailable.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If clustering deployment is applied with separate application nodes and data nodes, then system reliability is improved, but session queue congestion occurs when data node faults happen

Engineering Contradiction:
Improvesystem reliabilityVSAvoidservice request processing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements a feedback mechanism where the application node monitors the status of data nodes and sends evaluation response messages to the load balancer. When a data node fault is detected (through timeout or unresponsive status), the application node provides feedback that triggers the load balancer to adjust request distribution, preventing session queue congestion by stopping new requests from being routed to failed data nodes.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent applies preliminary action by having the load balancer proactively check the status of data nodes before routing service requests. The load balancer maintains status information about data nodes and evaluates their availability in advance, so that when a fault occurs, requests are already redirected to healthy nodes before congestion can occur in the session queue.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the application node processes service requests in order from the session queue, then processing reliability is improved, but response time increases when the queue is occupied

Engineering Contradiction:
Improveprocessing reliabilityVSAvoidresponse time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts the status monitoring and evaluation function from the application node's request processing workflow. By separating the fault detection mechanism (checking data node status) from the request processing queue, the system can quickly identify and respond to data node failures without being blocked by the sequential processing nature of the session queue, thereby reducing time loss while maintaining processing reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If the load balancer distributes service requests to multiple data nodes, then processing capacity is improved, but congestion occurs when some data nodes fail

Engineering Contradiction:
Improveprocessing capacityVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The feedback mechanism allows the system to dynamically adjust to data node failures. The load balancer receives status information from application nodes about data node health and automatically redistributes requests to healthy nodes. This feedback-driven approach maintains high processing capacity by utilizing available resources while managing the complexity of distributed system failures through automated response protocols.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3264723B1Method, related apparatus and system for processing service request
Publication Date: 2020.09.16 HUAWEI TECH CO LTD
  • EP3264723B1 patent drawingFigure 1~2
  • EP3264723B1 patent drawingFigure 3~4
  • EP3264723B1 patent drawingFigure 5~7

AI summary

The present invention provides a service request processing method, a related apparatus, and a system. In the service request processing method provided in embodiments of the present invention, after receiving a service request reported by a network element, a load balancer first determines whether status information of a data node that corresponds to the service request indicates that the data node is available; and rejects, if the status information indicates that the data node is unavailable, to deliver the service request to an application node. According to the embodiments of the present invention, the received service request may be temporarily stored, and the temporarily stored service request is delivered to an application node when the status information of the data node changes to that the data node is available, thereby avoiding a case, in the prior art, in which a service request is delivered to an application node when a status of a data node is unavailable, causing that a large quantity of service requests are congested in a session queue in the application node. Compared with the prior art, in the embodiments of the present invention, service requests for other normal data nodes can be processed in time, thereby improving overall efficiency of processing service requests.