Service-Based Forwarding Policy for Dynamic Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network systems are inflexible in managing service load balancing, leading to inefficiencies in resource allocation and client connectivity during varying demand periods, as they cannot dynamically distribute service sessions based on the capacities and capabilities of service load balancers, resulting in suboptimal performance and client disconnection issues.
Innovation Solution
Implementing a service-based policy in a data network that involves receiving data packets, matching them against classification rules based on the capabilities of servicing nodes, and dynamically generating forwarding policies to direct packets to the most suitable nodes, ensuring efficient resource utilization and client connectivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single service load balancer is used to handle software patch download demand, then the system is simple to manage, but it cannot handle dramatic increases in download demand during major software updates
Solution Approach 1:
The patent combines multiple service load balancers into a unified group that shares a common IP address. This allows the system to handle increased download demand by distributing traffic across multiple balancers while maintaining a single point of contact for clients, thus increasing productivity without proportionally increasing complexity.
Solution Approach 2:
The patent creates a universal IP address that can be shared across multiple service load balancers. This universal address serves as a common entry point for all clients regardless of which specific balancer handles their requests, enabling the system to scale capacity while maintaining simple client access.
2Productivity
If additional service load balancers with new IP addresses are added to handle increased demand, then the system can handle more traffic, but client devices must learn the new IP addresses before requesting service
Solution Approach 1:
The patent implements a universal IP address that is shared across multiple service load balancers. Clients only need to know this single universal address to access the service, regardless of how many balancers are in the group. This eliminates the need for clients to learn multiple IP addresses while still enabling the system to handle increased traffic through multiple balancers.
3Productivity
If service load balancers are removed after demand declines, then resources are optimized, but client devices that learned the new IP addresses can no longer use the patch service
Solution Approach 1:
The patent uses a universal IP address shared by the service load balancer group. When balancers are added or removed, the universal address remains the same, ensuring continuous client connectivity. The system can optimize resources by removing balancers after demand declines without affecting client access, as clients continue to use the unchanged universal address to reach the remaining balancers in the group.
Solution Approach 2:
The universal IP address acts as an intermediary between clients and the actual service load balancers. This intermediary abstracts the dynamic nature of the balancer group from clients, allowing the system to reconfigure its infrastructure (adding or removing balancers) without disrupting client connectivity. The network infrastructure transparently routes traffic to available balancers based on the universal address.
4Adaptability or versatility
If service load balancers are configured with different capabilities (memory, processing capacity, security processors), then the system can satisfy diverse client needs, but the network cannot dynamically distribute sessions based on these capabilities
Solution Approach 1:
The patent implements dynamic session distribution based on the capabilities of service load balancers. The system continuously monitors balancer capabilities (memory, processing capacity, security processors) and dynamically routes client sessions to the most appropriate balancer. This dynamic adaptation allows the system to satisfy diverse client needs while efficiently utilizing the heterogeneous capabilities of different balancers.
Solution Approach 2:
The patent applies local quality by matching specific client needs with the corresponding capabilities of individual service load balancers. Different balancers are optimized for different functions (e.g., security-processed balancers for encrypted sessions, high-memory balancers for large data transfers), and the system routes traffic to the appropriate local resource based on the specific service requirements.
Data Source
AI summary
Methods and systems are provided for forwarding data packets in a service network using a service-based policy. A gateway node may receive data packets from a service session and send the data packets for a forwarding node. The forwarding node may match the service session data packet against a service address and forwarding policy, the forwarding policy being based on capabilities and service configurations of the servicing nodes in the service network. Forwarding node may then forward the service session data packet to a servicing node in accordance with the forwarding policy, and send the service session data packet to a server for processing.