Enhanced Application-Aware Routing in Cisco SD-WAN
A Enhanced Application-Aware Routing is a Cisco Catalyst SD-WAN capability that
-
speeds detection of tunnel performance issues
-
enables faster traffic redirection to meet SLA thresholds, and
-
reduces switchover time compared to default behavior.
Role of BFD in Performance Monitoring
Bidirectional Forwarding Detection (BFD) detects link failures and collects performance routing (PfR) data for SD-WAN tunnels (IPsec and GRE).
Each BFD hello packet gathers:
-
Latency: Round-trip time between BFD echo request and reply.
-
Jitter: Variation in packet arrival delay (irregular timing of packets).
-
Loss: Number of echo requests that do not receive replies.
With the default BFD hello timer of 1 second, the system collects one PfR sample every second.
Poll Interval and Data Averaging
The system collects PfR samples throughout the poll interval (default: 10 minutes).
During each poll interval, it calculates the average for latency, jitter, and loss.
To make dynamic path decisions based on application-aware routing SLAs, the system uses a default multiplier of 6.
This multiplier evaluates multiple poll-interval averages before making routing decisions.
A poll interval average represents the time between consecutive data-collection events and indicates how frequently the system samples network performance.
Convergence Time Considerations
Convergence time is the time required for the network to recover and resume normal operation after a disruption.
-
Default detection of slowly degrading WAN circuits: 10 minutes to 1 hour
-
With the lowest recommended settings (2-minute poll interval × 6 intervals): 2 to 12 minutes
Using very low poll intervals can cause false positives and traffic instability due to insufficient performance data.
Without enhanced application-aware routing enabled, Cisco IOS XE Catalyst SD-WAN devices require several minutes to switch traffic from one network path to another to meet SLA requirements when the loss, latency, and jitter exceed specific threshold values. Enabling enhanced application-aware routing speeds the detection of tunnel performance issues. This enables Cisco IOS XE Catalyst SD-WAN device to redirect traffic away from tunnels that do not meet SLA requirements.
Pfr measurements
|
Metric |
Source |
Description |
|---|---|---|
|
Loss |
BFD |
Measured as loss of BFD packet at 1pps or one packet in n_app_probe_class (n-apc) sec If the application probe class (APC) configuration is not set, BFD packet loss occurs at a rate of 1 packet per second (1 pps). With the APC configuration, packet loss reduces to 1 packet every N seconds. For more information see, Application Probe Class. |
|
Latency |
BFD |
RTT measurements 1 pps or one packet in n-apc sec Without the application probe class (APC) configuration, RTT packet loss occurs at a rate of 1 packet per second (1 pps). With the APC configuration, packet loss reduces to 1 packet every N seconds. |
|
Jitter |
BFD |
Variation in RTT |
Application-aware routing design and measurement intervals
-
The default BFD hello-interval is 1 second, and the app-route/SLA poll-interval is 10 minutes.
-
The BFD hello-interval defines how often the BFD (Bidirectional Forwarding Detection) protocol sends hello packets to detect the liveliness of a network path. The app-route/SLA poll-interval defines how often the network monitoring system collects data or measures network metrics related to application routes or Service Level Agreements (SLAs).
-
By default, the system calculates to 60 minutes using 1 pps x 600 sec x 6 buckets. This calculation multiplies 1 packet per second (pps) by 600 seconds (10 minutes) and then by 6 buckets, resulting in a default poll-interval of 60 minutes.
-
Experts suggest using a poll-interval of 120 seconds (2 minutes) and a multiplier of 5, which produces a 10-minute interval. This recommendation helps achieve a specific monitoring frequency.
-
Reducing the poll-interval or multiplier improves detection time but can create false positives because PfR metrics rely on a small number of samples. Decreasing these values speeds up detection of network performance issues but increases the likelihood of false positives due to fewer data samples. You must balance detection time with the accuracy of PfR (Performance Routing) metrics.
-
To improve measurement accuracy at a faster rate, reduce the BFD hello-interval. Decreasing the interval at which hello packets are exchanged allows more frequent detection of network path liveliness and improves measurement accuracy.
Benefits of enhanced application-aware routing
-
Improved the PfR metrics: Inline data now improves PfR metric measurements (loss, latency, and jitter) by providing more accurate and detailed analysis. Cisco IOS XE Catalyst SD-WAN devices process and inspect this data directly at the network edge, enabling real-time inspection and decision-making.
-
Quick Enhanced-App-Route Detection and SLA Enforcement: Reduced the PfR poll interval to a minimum of 10 seconds to enhance App-Route detection and SLA enforcement. Cisco IOS XE Catalyst SD-WAN devices can now quickly detect circuit degradation and remove tunnels from SLA forwarding when thresholds are not met, ensuring efficient and reliable performance.
-
Speed: Increased the speed of SLA switch-over.
-
SLA Dampening: Introduced SLA dampening to enable smoother transitions back to SLA forwarding. Tunnels now undergo a dampening period before re-entering SLA forwarding to prevent disruptions and instability and to minimize negative effects on network performance.
-
Measurements: Enhanced measurement of loss, latency, and jitter.
Guidelines of enhanced application-aware routing
Guidelines of enhanced application-aware routing
-
Both GRE and IPSEC tunnels are supported.
-
All existing TLOCs and WAN interface types, including physical, sub interface, loopback bind, dialer, and LTE interfaces, are supported.
-
TLOC Extension tunnels are supported.
-
Both IPv4 and IPv6 underlay tunnels are supported.
-
SLA update and switchover occur at a minimum interval of 10 seconds.
-
Does not impact tunnel scale and maintains minimal impact on memory and performance.
-
Provides support with and without app-probe class configuration in SLA classes. For more information on app-probe class, see Application Probe Class.
-
SLA dampening is supported.
Compatibility with Cisco IOS XE catalyst SD-WAN devices without enhanced application-aware routing
-
In these scenarios:
-
On the local side: When the Cisco IOS XE Catalyst SD-WAN device is upgraded to Cisco IOS XE Catalyst SD-WAN Release 17.12.1a and later and has EAAR (Enhanced Application-Aware Routing) enabled.
-
On the remote side: When the Cisco IOS XE Catalyst SD-WAN device is not upgraded to Cisco IOS XE Catalyst SD-WAN Release 17.12.1a and the EAAR is not enabled.
It falls back to using BFD-based measurements where support for older releases or disabled features is required.
-
-
If both the local and remote sides are using Cisco IOS XE Catalyst SD-WAN Release 17.12.1a but the EAAR feature is not enabled, it reverts to using BFD-based measurements.
 Note |
The EAAR feature is disabled by default to support existing deployments. |
Restrictions for enhanced application-aware routing
Restrictions for enhanced application-aware routing
-
The branch device on which you enable this feature does not support loopback unbind mode. The loopback unbind mode refers to a network interface configuration in which the loopback device is disconnected from the network stack.
-
GRE tunnels do not support per-queue measurement. Per-queue measurement monitors and analyzes network traffic on a per-queue basis by measuring and collecting metrics and statistics for each individual queue in a network device or system. A queue acts as a buffer where packets are stored before transmission or processing.
Feedback