PoP Selection Controller for Dynamic SaaS Network Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing geo-centric approach to selecting Point of Presence (PoP) servers for cloud-based Software as a Service (SaaS) applications often results in suboptimal delivery due to factors beyond geographical proximity, such as network health metrics like packet loss, latency, and jitter, which can lead to unreliable and inefficient service delivery.
Innovation Solution
A computing system and method that employs a PoP Selection Controller (PSC) to determine network health metrics between PoP servers and network branches, as well as between PoP servers and SaaS host servers, to select the most suitable PoP server for each network branch based on composite health metrics, including packet loss, latency, and jitter, and maintains a lookup table for real-time ranking and updating of PoP servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If geo-centric PoP selection is used, then ease of operation is improved, but reliability deteriorates
Solution Approach 1:
The system implements feedback by continuously monitoring network health metrics (packet loss, latency, jitter) between PoP servers and both network branches and SaaS host servers. This real-time feedback enables dynamic adjustment of PoP selection decisions, ensuring that the chosen PoP provides reliable service delivery while maintaining operational simplicity through automated controller management.
Solution Approach 2:
The PoP selection transitions from a static geo-centric approach to a dynamic health-metric-based approach. The PoP selection controller continuously evaluates multiple PoP servers based on real-time network health metrics and dynamically selects the optimal PoP for each network branch, allowing the system to adapt to changing network conditions while maintaining ease of operation.
2Device complexity
If geo-centric PoP selection is used, then device complexity is reduced, but productivity deteriorates
Solution Approach 1:
The PoP selection controller acts as an intermediary that manages the complexity of evaluating multiple network health metrics and PoP servers. It automatically performs the complex task of monitoring packet loss, latency, and jitter across multiple PoP connections and selecting the optimal PoP, thereby improving application delivery speed without requiring network branches to directly manage the complexity of PoP selection criteria.
Solution Approach 2:
The system performs preliminary actions by pre-establishing health metric monitoring for multiple PoP servers and maintaining a ready-to-use selection framework. This allows rapid PoP selection when needed, improving productivity by having evaluation results prepared in advance rather than computing selections in real-time during application delivery.
3Reliability
If health metric monitoring is implemented, then reliability is improved, but use of energy increases
Solution Approach 1:
The system implements partial monitoring by focusing health metric collection on critical paths - specifically between PoP servers and network branches, and between PoP servers and SaaS host servers. Rather than monitoring all possible network parameters universally, the system selectively monitors the specific metrics (packet loss, latency, jitter) that directly impact service delivery reliability, reducing unnecessary energy consumption while maintaining high reliability.
Data Source
AI summary
A computing system may include point of presence (PoP) servers coupled to a wide area network (WAN) and configured to receive client requests for a Software as a service (SaaS) application(s) from different network branches coupled to the WAN, and connect the network branches with a given SaaS host server from among different SaaS host servers coupled to the WAN and providing the SaaS application(s). The system may also include a PoP selection controller (PSC) coupled to the WAN and cooperating with the PoP servers to determine first network health metrics for connections between the PoP servers and the network branches, determine second network health metrics for connections between the PoP servers and the SaaS host servers, and select a respective PoP server for each network branch to be connected with for providing the SaaS application(s) based upon the first and second network health metrics.


