Service Packet Scheduling by Network Service Level Labels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud service solutions fail to provide optimal network quality for customers due to congestion or insufficient network capacity, resulting in degraded experience for high-level services when all services are routed through a single, potentially congested line.
Innovation Solution
A service scheduling method that configures network resources based on required service levels by matching target level labels with routing tables to direct service packets through optimal network paths, ensuring high-level services utilize high-quality resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all services are provided through a single candidate line with better network quality, then network reliability is improved, but network performance deteriorates due to congestion and insufficient capacity
Solution Approach 1:
The patent segments services into different levels (high-level and low-level services) and routes them through different network lines. High-level services are routed through a first network line while low-level services use a second network line, allowing simultaneous optimization for both reliability and performance without requiring a single line to handle all traffic
Solution Approach 2:
The patent implements dynamic service scheduling that automatically adjusts routing based on real-time network conditions. When the primary network line becomes congested, the system dynamically redirects high-level services to alternative lines, maintaining both reliability and performance through adaptive resource allocation
2Extent of automation
If a BGP route selection rule is used to determine communication routes, then routing automation is improved, but network quality deteriorates because the determined route may not be optimal
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors network quality metrics (latency, packet loss, bandwidth) and uses this information to dynamically adjust service routing decisions. This feedback loop enables the system to maintain automated routing while ensuring optimal network quality by redirecting services away from congested or degraded lines
Solution Approach 2:
The patent changes the routing parameters from static BGP path selection to dynamic quality-based selection. By introducing real-time network quality parameters (latency, jitter, packet loss) as decision criteria and adjusting service routing based on these parameter changes, the system achieves both automation and optimal network quality
3Quantity of substance
If multiple network lines are available for different services, then network capacity is improved, but device complexity increases due to the need for service scheduling and routing management
Solution Approach 1:
The patent applies local quality by assigning different service levels (high-level and low-level) to different network lines based on their specific characteristics. Each line is optimized for its designated service type, with high-level services receiving premium routing and low-level services using standard routing, thereby managing complexity through specialized rather than generalized routing logic
Data Source
AI summary
This application discloses a service scheduling method and a device. The method includes: receiving service configuration information input by a user, where the service configuration information includes a network service level required by a service; obtaining a first service packet, where the first service packet carries content and a target level label of the service, and the target level label is used to identify the network service level required by the service; then matching the target level label with a plurality of level labels in a first routing table; when the target level label matches any one of the plurality of level labels, determining a target port label based on the level label obtained through matching; and finally, controlling a point of presence to send the first service packet from a target port indicated by the target port label to a client.