As part of our IWAN series, I wanted to provide a deeper dive into PfR. Why PfR? It is a fundamental feature that helps customers protect critical apps while increasing bandwidth utilization. I think it is fair to say, every organization can benefit tremendously from this powerful capability.
PfR or Performance Routing is a feature that complements traditional IP routing protocols by adding application intelligence when making routing decisions. Why do we need application intelligence? Routers forward data packets based on their routing tables which are built using dynamic routing protocols such as RIP and OSPF to calculate the shortest path to the destination for the data packets. RIP and OSPF do not look into data packets to determine the type of application they belong to when making routing decisions. As a result if the application is time sensitive like voice over IP (VOIP) or bandwidth intensive like a file backup data packets are treated with the same priority and will be sent over the same route until they reach their destination. This can create problems if you have a single WAN link since a file backup could consume all bandwidth preventing voice packets from passing in a timely manner and impacting the quality of the voice call. QoS or Quality of Service can help to prioritize data on a single link but you may ultimately need more bandwidth.
For the majority of customers each branch location will typically have two WAN links. The first is the primary, and the second is for backup if the primary link fails. Unfortunately the backup link sits idle when the primary is active mainly because having a second link between the same end-points can create loops or cause applications to send packets over both links, which can ultimately negatively impact application performance. So to reduce cost and improve redundancy customers may purchase the backup link from a different service provider, accept a lower SLA or even a lower speed. So how can you take advantage of your backup ink when your primary link is active? This is where PfR can help.By enabling PfR on both ends of your WAN links you can do the following:
- Maximize throughput by keeping both of your WAN links active simultaneously.
- Lower WAN costs by sending application data over cheaper links.
- Improve application performance by monitoring WAN link quality.
- Improve application reliability by redirecting traffic in the event of WAN link failure.
- Move VPN’s transparently between WAN links.
Depending on your network design you may need some or all of the above features. As more employees move to the branch and drive more traffic over the WAN, increasing throughput while maintaining application performance and business continuity is a top priority for IT and PfR can go a long way to alleviating those pain points.
If you want to see PfR in action, view the video demonstration provided by Cisco technical expert, Bill Reilly. You will also be happy to know that the current version of PfR (version 2) has many new enhancements to simplify your deployment – with target discovery of devices and configuration reduced down to 10 lines. Plus, it now scales to 500 sites.
PfR is currently supported on the Cisco ISR G2 and ASR1000 platforms. If you are refreshing your branch offices, you will want to consider the ISR-AX routers, which includes PfR and many other critical application and security services. For those of you that are aggregating a large number of WAN connections and require higher throughput Cisco has some new ASR bundles with dual 10G connectivity at a competitive price. The updated ASR1000 ordering guide can be found here.