Client-Driven Cloud Network Access via Aggregated Data Packets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional networks, such as 5G cellular networks, lack runtime state management to identify optimal settings and configurations for improved performance on client devices, leading to static settings that do not adapt to changing workloads during application runtime.
Innovation Solution
A client-driven cloud network access system that generates an aggregated data packet including a request, requested performance level, and application telemetry data, which is transmitted to multiple servers to select the best server for providing the service based on advertised performance levels, enabling dynamic optimization and affinity between applications and services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional top-down network approach is used, then network configuration is simplified, but performance parameters cannot adapt to changing client application needs
Solution Approach 1:
The patent inverts the conventional top-down configuration approach by implementing a bottom-up model where client applications autonomously select and configure network services based on their runtime performance needs. Instead of the network pushing configurations down to clients, applications actively query and select from available network services, fundamentally reversing the control flow to achieve adaptive performance optimization.
Solution Approach 2:
The system implements dynamic service selection where applications continuously monitor their performance requirements and adaptively choose network services based on changing workload conditions. The service selection process is not static but dynamically adjusts to runtime application behavior, enabling the network configuration to evolve alongside application needs without manual intervention.
2Productivity
If static network settings are provided, then system complexity is reduced, but performance cannot be optimized for varying workloads
Solution Approach 1:
The patent enables applications to self-select appropriate network services based on their own performance requirements without external intervention. Applications autonomously query available services, evaluate their capabilities, and select the most suitable service for their current workload, eliminating the need for complex centralized performance optimization mechanisms while achieving productivity improvements.
Solution Approach 2:
The system incorporates feedback loops where applications monitor their performance metrics and use this information to make informed service selection decisions. The runtime state management captures application behavior patterns and feeds this information back into the service selection process, enabling continuous performance optimization based on actual workload conditions rather than static configurations.
3Measurement precision
If multiple servers are queried for service selection, then service matching accuracy is improved, but network communication overhead increases
Solution Approach 1:
The patent implements preliminary action by having applications query multiple network services in advance to build a knowledge base of available services and their capabilities. This pre-querying phase allows applications to understand the service landscape before actual service selection is needed, reducing the communication overhead during runtime when performance-critical decisions must be made quickly based on pre-gathered information.
Data Source
AI summary
Embodiments of systems and methods for a client driven cloud network access system and method are described. In an illustrative, non-limiting embodiment, an IHS may include computer-executable instructions for receiving a request to perform an operation provided by a cloud service from an application, and generating an aggregated data packet including the request, a requested performance level to be provided by the service, and application telemetry data associated with a measured performance level of the application. The aggregated data packet is then transmitted to a plurality of servers that are each configured to provide the cloud service. Each of the servers generates and sends a response to the aggregated data packet that includes an advertised performance level at which the server can provide the cloud service. The IHS may then select one of the servers according to the advertised performance level, and communicate with the selected server to provide the requested operation.