SCP Service Instance Feedback for Targeted Request Throttling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In existing communication networks, the client node is unable to perform targeted throttling during indirect communication model D due to lack of knowledge about the selected service instances, leading to arbitrary throttling and looped error scenarios without specific overload information, which affects service request handling efficiency.
Innovation Solution
The introduction of a service communication proxy (SCP) that provides service information to the client node, including indications of available service instances and estimated throttling, allowing the client node to perform application-aware throttling based on overload information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If indirect communication model D is used with SCP node, then service request routing capabilities are improved, but client node cannot perform targeted throttling due to lack of service instance information
Solution Approach 1:
The SCP node acts as an intermediary that receives service requests from client nodes, performs discovery and selection to identify suitable service instances, and forwards requests accordingly. The SCP node mediates between the client node and service instances, enabling indirect communication while maintaining routing capabilities.
Solution Approach 2:
The system implements feedback by having the SCP node provide information about service instances back to the client node. This feedback mechanism allows the client node to receive overload information and perform targeted throttling based on the current state of service instances, creating a closed-loop control system.
2Measurement precision
If SCP node performs discovery and selection, then service instance selection is improved, but arbitrary throttling and looped errors occur without client node awareness
Solution Approach 1:
The SCP node provides feedback to the client node about service instance status, including overload information. This enables the client node to make informed throttling decisions rather than performing arbitrary throttling, and breaks looped error scenarios by providing visibility into the selection process and current service instance states.
Solution Approach 2:
The client node performs self-service by autonomously deciding when to throttle based on overload information received from the SCP node. Rather than relying on the SCP node to perform all throttling decisions, the client node uses the provided information to make intelligent throttling decisions at the source.
3Object-affected harmful factors
If client node performs throttling without service instance knowledge, then network congestion mitigation is attempted, but inefficient arbitrary throttling occurs
Solution Approach 1:
The SCP node provides feedback about service instance overload status to the client node, enabling informed throttling decisions. This feedback mechanism allows the client node to throttle only when necessary and only towards affected service instances, rather than performing arbitrary throttling that reduces overall efficiency.
Solution Approach 2:
The system changes the parameter of throttling from arbitrary to targeted by providing the client node with specific information about which service instances are overloaded. This parameter change enables precise control over throttling behavior, mitigating congestion only where needed while maintaining efficiency for healthy service instances.
Data Source
Figure 1
Figure 2
Figure 2
AI summary
A first service communication proxy, SCP node in a communications network, wherein the first SCP node is configured to operate as a SCP between a client node and one or more server nodes operating service instances of a first service. The first SCP node comprises: a memory comprising instruction data representing a set of instructions; and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to: receive a first service request from the client node, requesting the first service; and send service information to the client node, the service information comprising an indication of the service instances of the first service that are operating on the one or more server nodes.