Why Intelligent Traffic Steering is Critical for Performance and Cost Optimization

The challenge: static routing in a dynamic world

Traditionally, DNS-based routing strategies like round-robin or proximity-based decisions have been used to direct user traffic. However, these methods don't account for real-time performance metrics. A nearby server might be under heavy load or facing regional ISP issues, yet static routing would still direct users there, resulting in higher latency and degraded experience.

This issue is magnified in multi-cloud and multi-CDN environments, where routing inefficiencies can also drive up cloud and infrastructure costs. Without real-time visibility, you’re essentially flying blind.

The image above provides a great example of this challenge in a multi-cloud environment using AWS and Oracle. The global response time maps show noticeable regional performance differences between the two providers. Yet, because traffic is being routed using a static round-robin method (seen in the 52% AWS / 48% Oracle split), users in underperforming regions are still being directed to slower endpoints. For instance, the Oracle path shows significant latency in North America and Oceania, while AWS performs better there. Still, the lack of performance-aware steering results in an inefficient and inconsistent user experience.

The Solution: Adaptive traffic steering with Catchpoint and IBM NS1

By combining Catchpoint’s extensive Global Agent Network—spanning 2,880+ intelligent agents and millions of connected devices—with IBM NS1’s real-time DNS traffic steering capabilities, organizations can shift from reactive to proactive, performance-driven routing.

Check out the video below for an overview of how Catchpoint and IBM NS1 work together to enable real-time, performance-aware traffic steering.  

Catchpoint IPM continuously collects metrics such as DNS resolution time, connect time, SSL, wait time, and total response time using synthetic tests from backbone, last mile, cloud, and wireless networks. This real-time telemetry is then fed into NS1's filter chain logic.

NS1 applies this data through its intelligent filter chains, which support decisions based on geolocation, ASN, availability, and actual performance. The diagram below illustrates how the end user’s DNS request is routed through NS1, where IPM-powered performance metrics guide the decision-making engine. This ensures traffic is not just sent to the closest server, but the best-performing server at that moment.

This feedback loop enables true adaptive routing, drastically reducing latency and improving reliability without waiting for end-user impact to trigger change.  

Feeding Catchpoint Data into IBM NS1 Connect

Catchpoint enables real-time performance telemetry to be pushed directly into NS1 IBM Connect using Data Webhooks. This integration empowers traffic steering logic to be based on real, actionable insights rather than static assumptions.

Metrics such as DNS time, connect time, SSL handshake duration, wait time (time to first byte), TTFB, and overall response time are just the start. Users can also feed in custom data sources like CDN-specific server timing metrics or application-specific performance KPIs. This flexibility ensures that the traffic steering logic is tailor-fit to business and technical requirements whether optimizing for speed, reliability, or cost.

All this data is consumed by NS1 Pulsar's filter chain engine, which uses it to dynamically apply routing rules. These filters can be chained together based on geography, ASN, latency, availability, or any of the performance metrics provided by Catchpoint IPM.

As a result, routing decisions aren't just smart they're precisely aligned with your most up-to-date performance landscape.

Real-world impact: Better performance and lower costs

In our joint implementation example, traffic was initially split evenly between AWS and Oracle cloud environments. After enabling dynamic routing with IPM-fed data, 77% of the traffic shifted to the better-performing cloud region, significantly reducing wait times and improving end-user experience.

The results speak for themselves:

• 86.57% reduction in wait time on AWS

• 85.30% reduction in wait time on Oracle

Wait time improvements after enabling performance-based traffic steering

These improvements not only enhance digital experience but also minimize the frustration and churn that come with sluggish performance.

Beyond latency improvements, traffic steering also drives operational efficiency. By understanding which cloud or PoP performs best in specific regions, businesses can:

• Scale down underperforming infrastructure

• Optimize cloud costs by avoiding overprovisioning

• Improve regional performance by steering traffic intelligently

This is the power of proactive, performance-aware traffic steering.

Proactive, not reactive

Most traditional monitoring tools rely on real user monitoring (RUM) data, which only becomes useful after users have already experienced performance issues. Catchpoint IPM flips this model by using synthetic testing for proactive insights. This means traffic can be rerouted before users are impacted, creating a continuous feedback loop that enhances reliability and experience.

Next up: the setup guide

In an upcoming post, we’ll walk through the step-by-step implementation of intelligent traffic steering—from configuring Catchpoint tests to feeding metrics into NS1 and setting up intelligent filter chains. Watch this space.  

Learn More

• Check out the IBM NS1 Connect Documentation to get started with traffic steering and DNS configuration.  

‍