Service Request Management via Direct Server Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing client-server computing environments face inefficiencies in load balancing due to the overhead of central control mechanisms and out-of-band communication methods, which affect server and hardware communication processing component loads, leading to suboptimal service quality.
Innovation Solution
Implementing a system where a client device initially routes a request through a communication processing component using a load balancing algorithm, receiving a service instance ID and connection lease, and subsequently sends requests directly to the server for the duration of the lease, bypassing the communication processing component, thereby reducing overhead and improving load distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If a server locator service (SLS) is used to handle client requests through out-of-band communication, then server location capability is improved, but communication overhead increases due to additional discovery packets
Solution Approach 1:
The patent extracts the server location function from the data communication path by using out-of-band communication channels. The SLS runs on separate communication channels that do not carry client data requests, allowing server discovery to occur independently without interfering with or adding overhead to the data transmission path.
Solution Approach 2:
The SLS acts as an intermediary service that mediates between client devices and server devices. It receives discovery requests from clients, locates appropriate servers, and returns server information without the client needing to directly query or communicate with servers for discovery purposes.
2Productivity
If a central control mechanism is used for load balancing, then workload distribution is improved, but device complexity increases due to dedicated hardware controllers
Solution Approach 1:
The patent implements self-service load balancing where each server device independently monitors its own workload conditions and makes decisions about accepting or rejecting client requests. Servers autonomously participate in load balancing by responding to discovery requests with their current load status, eliminating the need for centralized control hardware.
Solution Approach 2:
The SLS is implemented as a software service that can run on any server device in the network, making the load balancing capability universal and not dependent on specialized hardware. Any server can potentially host the SLS and participate in the load balancing process, distributing the control function across multiple nodes rather than requiring a dedicated controller.
Data Source
AI summary
A computer-implemented system and method for managing service requests are disclosed. The system includes a service provider, having a number of server devices, that provides a service response in response to a client service request. The service response includes a service instance ID associated with a server and a connection lease. A communication processing component routes a first client service request to one of the servers. Subsequent client service requests and responses are transmitted directly between the client device and the server device associated with service instance ID included in the first service response for the duration of the connection lease.