Cloud Edge Network Tracing via Correlation Identifiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing synthetic probing technologies cannot provide full end-to-end network visibility due to the presence of cloud edge networks, which act as proxies, resulting in a visibility blind-spot between the cloud edge network and the application origin server, limiting the ability to identify network performance issues.
Innovation Solution
The implementation of telemetry data correlation logic in the cloud or datacenter environment enables path discovery and dynamic monitoring of network segments between the cloud edge network and the datacenter, using correlation identifiers to aggregate telemetry data and generate an end-to-end distributed network trace, thereby bridging the visibility gap.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If cloud edge networks act as proxies to improve application performance, then application delivery speed is improved, but end-to-end network visibility is lost
Solution Approach 1:
The patent introduces synthetic agents as intermediary components deployed both at the origin and edge locations. These agents act as mediators that can observe and measure network traffic without disrupting the proxy function of the edge network. The agents capture telemetry data including correlation identifiers that allow tracking of requests through the proxy chain, thus restoring visibility while maintaining the performance benefits of edge networking.
Solution Approach 2:
The patent segments the network monitoring function by deploying separate synthetic agents at different locations (origin and edge). Rather than attempting single-point monitoring that would be blocked by proxies, the system divides monitoring into multiple segments that collectively provide end-to-end visibility. Each agent independently captures local telemetry and correlates it with other segments using correlation identifiers.
2Loss of time
If edge services are deployed remote from origin server to improve performance, then application response time is improved, but network path visibility is reduced
Solution Approach 1:
Synthetic agents serve as mediators that bridge the visibility gap created by remote edge deployment. By placing agents at both origin and edge locations, the system can measure network path characteristics without requiring direct visibility between geographically separated components. The agents correlate measurements using identifiers embedded in traffic, enabling path detection despite physical separation.
Solution Approach 2:
The system implements feedback mechanisms where synthetic agents continuously monitor and report network conditions back to the monitoring system. Telemetry data including latency, packet loss, and path information are collected and correlated to provide ongoing visibility into network performance. This feedback loop enables dynamic adjustment and continuous measurement of network paths between origin and edge.
3Ease of operation
If cloud edge networks are used to optimize application delivery, then user experience is improved, but troubleshooting capability is limited
Solution Approach 1:
Synthetic agents act as intermediaries that enable troubleshooting by providing observability into the edge network path. These agents capture detailed telemetry data about network conditions, latency, and traffic flow without interfering with user experience. The collected data can be analyzed to identify performance issues and guide troubleshooting efforts while users continue to benefit from optimized delivery.
Solution Approach 2:
The system performs preliminary monitoring and baseline establishment through synthetic agents before problems occur. By continuously collecting telemetry data and establishing normal performance baselines, the system can quickly detect deviations and identify issues when they arise. This preliminary action enables faster troubleshooting and resolution while maintaining optimal user experience during normal operation.
Data Source
AI summary
Techniques are described for generating an end-to-end distributed network trace involving cloud edge networks. In one example, a cloud or datacenter environment obtains, from an edge node in a cloud edge network, one or more network communications that include a correlation identifier associated with the one or more network communications and an identifier of the edge node. Based on the identifier of the edge node, the cloud or datacenter environment provides a network probe to the edge node. The cloud or datacenter environment obtains, from the edge node, telemetry data that is generated responsive to the network probe. The cloud or datacenter environment provides the telemetry data and the correlation identifier to an aggregation server that is configured to, based on the correlation identifier, aggregate the telemetry data with further telemetry data to generate an end-to-end distributed network trace associated with the one or more network communications.