Load Balancing Module Allocating Requests via Server Capacity Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing techniques face challenges in efficiently managing processing requests across servers, particularly in handling server resource availability and network congestion, which can lead to request rejection and increased risk of Denial of Service attacks.
Innovation Solution
A method and apparatus that utilize a load-balancing module and a load-monitoring module to interpret server processing capacity signals through existing communications protocol fields, such as the ECN field, to dynamically allocate processing requests based on server capacity, allowing servers to signal their resource availability and enabling informed load balancing decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional load balancing methods (round robin, random, static schemes) are used to distribute requests, then implementation is simple and device complexity is low, but server resource utilization is inefficient and responsiveness is reduced
Solution Approach 1:
The patent implements feedback mechanisms where servers send processing responses containing capacity signals back to the load-balancing module. The load-monitoring module analyzes these responses and dynamically adjusts request allocation based on current server capacity, creating a closed-loop control system that optimizes resource utilization while maintaining adaptive complexity only where needed.
Solution Approach 2:
The system performs preliminary actions by having servers proactively signal their processing capacity in advance through processing responses before receiving new requests. This allows the load-balancing module to make informed allocation decisions based on pre-reported capacity status rather than reacting to overload conditions after they occur.
2Productivity
If dynamic load balancing schemes that monitor server conditions are implemented, then server resource allocation is optimized and productivity improves, but device complexity and monitoring overhead increase
Solution Approach 1:
The patent makes the processing response serve multiple functions: it conveys the actual processing result from server to client and simultaneously carries capacity signaling information for load balancing decisions. This multi-functionality eliminates the need for separate monitoring messages, reducing communication overhead and system complexity while maintaining dynamic allocation efficiency.
Solution Approach 2:
Servers automatically generate and send processing responses with embedded capacity signals without requiring external probing or monitoring infrastructure. The load-monitoring module simply analyzes these self-generated responses, eliminating the need for complex active monitoring systems while achieving optimized request allocation based on real server capacity.
3Reliability
If load balancers send test requests (ping requests) to each server to check availability, then server capacity information is obtained, but network traffic increases and processing time is consumed
Solution Approach 1:
The processing response serves dual purposes: delivering the actual service result to the client and providing capacity status information for load balancing. This eliminates the need for separate test requests entirely, as the same communication channel carries both service data and capacity signaling, saving time and reducing network traffic.
Solution Approach 2:
The patent extracts the capacity information need from the service response and utilizes it for load balancing decisions without requiring additional service interactions. By taking out the capacity checking function from separate probe requests and integrating it into the existing service response flow, the system obtains reliability information without the time penalty of additional requests.
4Reliability
If servers reject requests when short of resources, then server overload is prevented and reliability is maintained, but request throughput decreases and productivity is reduced
Solution Approach 1:
Servers proactively signal their capacity status in processing responses before becoming overloaded. The load-balancing module uses this advance information to redirect future requests to appropriately capacity-equipped servers, preventing overload conditions rather than reacting to them. This maintains server stability while maximizing throughput by keeping requests flowing to capable servers.
Solution Approach 2:
The system implements continuous feedback loops where servers report capacity status and the load-balancing module dynamically adjusts request routing. This feedback mechanism prevents server rejection of requests by proactively balancing load before capacity thresholds are exceeded, maintaining both server stability and high request throughput through adaptive real-time control.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and apparatus are disclosed for allocating processing requests between a plurality of servers (16a, 16b, 16i), each processing request being a request from a client processor (10a, 10b, 10n) for processing to be performed by a server. The method and apparatus involve a load-balancing module (13) and an associated load-monitoring module (14) interacting with each other and with the plurality of servers. The load-monitoring module (14) is configured to receive processing responses from the servers, identify codepoints in predetermined fields thereof, and to interpret them according to a server processing capacity signalling protocol. The load-balancing module (13) is configured to receive processing requests submitted by the client processors and to forward them for processing at a server selected in dependence on the interpretation according to the server processing capacity signalling protocol of the codepoints in the predetermined fields of processing responses previously received from the respective servers.