SaaS Probe Aggregation for Network Path Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for probing Software-as-a-Service (SaaS) clouds result in redundant probes and high volumes of network traffic due to each router independently monitoring all possible paths, leading to inefficient resource utilization and performance issues.
Innovation Solution
Distribute probing responsibilities among multiple routers in a colocation facility and share the context of probe results among the routers, allowing each router to determine the best path to a SaaS cloud without individually probing each path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If each router independently probes all possible paths to SaaS clouds, then routing performance data can be obtained for path selection, but redundant probes are generated and network traffic volume increases significantly
Solution Approach 1:
The patent segments the probing function by assigning different SaaS application probes to different routers within the same site. Instead of every router probing every application, each router is responsible for probing a specific subset of applications, thereby eliminating redundant probes while ensuring that routing performance data is collected for all applications across the distributed router system.
2Reliability
If multiple routers in the same site use the same ISP and probe the same applications, then comprehensive routing performance data can be collected, but identical probe results are generated across all routers
Solution Approach 1:
The patent merges the probing efforts of multiple routers by having them share probe results through a distributed data structure. When one router performs probes for a particular SaaS application, the results are made available to other routers in the same site, eliminating duplicate probing activities while maintaining data completeness across the system.
Solution Approach 2:
The patent creates a universal probe result sharing mechanism where a single probe execution by any router in the site benefits all routers. The distributed data structure allows probe results to be universally accessed across the router cluster, enabling each router to make informed routing decisions without independently performing redundant probes.
3Measurement precision
If probing is enabled at per device per interface level for all possible paths, then detailed routing performance monitoring is achieved, but resource utilization becomes inefficient at scale
Solution Approach 1:
The patent segments the probing workload across multiple routers rather than concentrating it at the interface level of each individual device. By distributing the probing responsibility and allowing routers to share results, the system maintains detailed monitoring capabilities while reducing the resource burden on each individual router interface.
Data Source
AI summary
Techniques for determining an optimal path for data traffic are described. The techniques may include receiving, by a first router located at a remote location of an enterprise organization and from a second router located at the remote location, first routing performance data indicating a first performance of a first path form the remote location to an application infrastructure. The first router also receives second routing performance data indicating a second performance of a second path from the remote location to the application infrastructure from a third router. The first router compares the first routing performance data and the second routing performance data, and based on the comparing, determines whether the first path or the second path is optimal, and transmits data traffic to the application infrastructure via the optimal path.