Global Load Balancing for Heterogeneous Devices via Metric Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing systems face challenges in effectively managing heterogenous devices across different geographic regions and networks, as they often lack standardized status information, leading to suboptimal load distribution and decision-making.
Innovation Solution
A load balancing appliance that collects metrics from heterogenous devices using SNMP and a metric exchange protocol, allowing users to configure global load balancing based on diverse metrics, enabling flexible and informed decision-making across different types of local load balancers and servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple types of local load balancers from different manufacturers are deployed across geographic regions, then the system can serve more users and distribute load across more devices, but the devices cannot directly provide standardized status information to the global load balancer
Solution Approach 1:
The patent introduces an intermediary component that translates and normalizes status information from multiple heterogeneous load balancer devices into a standardized format that the global load balancer can process. This intermediary layer enables communication between devices with different protocols and formats without requiring changes to the existing diverse device infrastructure.
Solution Approach 2:
The system transforms status information parameters from various sources into a unified parameter set. By changing the representation format of status information from device-specific parameters to standardized parameters, the global load balancer can effectively compare and evaluate load conditions across heterogeneous devices.
2Adaptability or versatility
If different types of local load balancers with different metrics are used, then the system gains flexibility in load distribution strategies, but the global load balancer cannot readily access and compare different parameters for decision-making
Solution Approach 1:
The patent creates a universal interface that can handle multiple types of metrics and parameters from different load balancer devices. This universal mechanism allows the global load balancer to access and compare diverse parameters (such as response time, throughput, error rates) from heterogeneous devices through a single standardized access point, maintaining both flexibility and ease of operation.
3Measurement precision
If a global load balancer monitors requests and responses between clients and servers to determine server load, then accurate load information can be obtained, but this requires direct monitoring capability across all devices which heterogeneous devices cannot provide
Solution Approach 1:
The patent introduces an intermediary that collects and standardizes load information from heterogeneous devices, enabling the global load balancer to obtain accurate measurements without requiring direct monitoring capabilities across all device types. This intermediary handles the complexity of diverse monitoring protocols and formats.
Solution Approach 2:
The system implements a feedback mechanism where status information from load balancers and servers is continuously collected, standardized, and fed back to the global load balancer for decision-making. This feedback loop ensures accurate and up-to-date load information is available while the intermediary handles the complexity of diverse data sources.
Data Source
AI summary
The present invention provides improvements to load balancing by providing a load balancing solution that distributes a load among a plurality of heterogenous devices, such as different types of local load balancers, using metrics collected from the different devices. The load balancing appliance collects metrics from heterogenous devices using a network management protocol and communication model, such as a Simple Network Management Protocol (SNMP). These heterogenous device metrics are available on the load balancing appliance with appliance determined metrics and metrics obtained by the appliance from homogenous devices using a metric exchange protocol. Via a configuration interface of the appliance, a user can select one or more of these different metrics for global load balancing. As such, the load balancing appliance described herein obtains a multitude of metrics from the different devices under management. Additionally, the load balancing appliance described herein provides great flexibility in allowing the user to configure the global load balancer based on the user's understanding of these multitudes of metrics and to take into account the different characteristics and behaviors of the heterogenous devices.