Have an account?

  •   Personalized content
  •   Your products and support

Need an account?

Create an account

Addressing SD-Core: Boosting SD-WAN performance without MPLS

SD-WAN has emerged as a cost-effective way to connect branches to a corporate network. But enterprises need to address middle-mile performance issues at the core of the network.

SD-WAN technology has emerged as an important means to save money and boost networking infrastructure flexibility.

But as with other interconnected technologies, SD-WAN technology relies on other architecture to garner its maximum benefits. In this piece, we examine how SD-WAN architecture can improve your infrastructure, but only when used in conjunction with other private networks, like Multiprotocol Label Switching (MPLS) for high latency paths.

Software-defined wide area network (SD-WAN) technology enables branch offices to connect to a corporate WAN or multiple data center locations over large geographic distances, using any type of transport. SD-WAN abstracts much of the network connectivity management into the software layer.

SD-WAN technology reduces costs, increases network agility, and improves reliability by exploiting affordable Internet-based services. But architectures that lack a global WAN structured around MPLS —a protocol designed to speed traffic flows—can introduce consistency issues. New software-defined core (SD-Core) architectures offer a solution, providing enterprises with high-quality, affordable backbone alternatives to traditional MPLS services. Here we look at how to boost SD-WAN performance without relying on MPLS.

The Internet core problem

First we must understand the glue binding together this complex mesh of IP networks we call the Internet: the provider relationships guiding how they pass traffic between one another. Border Gateway Protocol (BGP)—a protocol that optimizes traffic flows—is a marvel in being able to translate those relationships into the world of Internet peering. But for all of its genius, BGP can’t overcome provider business interests. IP networks have peering relationships—the arrangement of traffic exchange between Internet service providers—that cost money. Thus, the more direct, low-latency path, may not be the path selected in every case. Peering points can get congested, which creates variation in latency and packet loss.

With MPLS, providers were more accountable for customer experience. They have a business interest in minimizing latency and optimizing routing to meet an SLA. Failure to do so could increase customer dissatisfaction, customer churn and lost revenue.  

But Internet-backbone providers seek to maximize the value of their networks not the performance of any single application. Often it can make more sense to dump traffic off on another provider’s backbone than take it across a faster route along their own network. This is how you end up with “hot potato routing,” a phenomenon all too familiar to Internet engineers.

Much of the issues with Internet routing occur in the core of the network. When traffic is kept within region, the impact of the Internet core is often minimized. A 20% difference on a 20ms path is insignificant for most applications. But the same variation on a 200ms path can mean the difference between a clear voice call and an unintelligible one.

SD-WAN Experts’ recent testing highlighted the problems of the Internet core. We measured and compared the end-to-end delay among several last mile services, several Internet backbones, and a private backbone. Our testing showed that while as a percentage, last-mile connections might be the most erratic, the sheer length across the Internet core in a global connection makes the middle-mile performance a far greater determiner of overall latency.  

The last mile variation for four paths to Bangalore from San Jose, London, Tokyo and Sydney was 5.88ms (3ms was the median). By contrast, the middle miles varied from 36% to 85%—92ms to 125ms—a 20x greater impact on the connection.

SD-Core architectures

If enterprises want to eliminate costly MPLS, they need to address these middle-mile performance issues of the Internet core. SD-WAN appliances alone aren’t the answer. An alternative backbone is needed. One that’s more affordable than MPLS, more consistent than the Internet, and, unlike carrier-managed SD-WAN services, does not lock enterprises into one provider’s access network. Three such approaches to an SD-Core are now available and use local Internet for access.

  1. Independent MPLS backbones. With independent MPLS backbones, a service provider builds a global, MPLS core network with its own SD-WAN edge device at the customer premises.

    The L2 MPLS network provides great performance and the use of Internet access makes this approach more affordable and flexible than a classic MPLS service. But pricing is still higher than other approaches, a fact that stems in part from the hardware-centric nature of building out an MPLS Core network. Customers are also limited to the SD-WAN edge offering of the independent MPLS backbone provider.

  2. Software-defined backbones. By contrast, software-defined backbones build overlays across existing IP backbones. Route selection is done using a distributed, optimal feedback control system.

    By exploiting IP, cost savings for core bandwidth are more significant than with MPLS backbone. Software-defined backbones also do not lock enterprises into their edge hardware. Companies are free to use any SD-WAN device (or any other device) that can build the necessary IPsec tunnels to the backbone POP, which is within 20ms of most commercial centers.

  3. Cloud-based SD-WAN. With cloud-based SD-WANs, a cloud-scale software stack running in a provider’s point-of-presence (PoP) executes most SD-WAN and security functions. The provider’s edge devices are very “thin,” with only enough functionality to bring traffic into the cloud-based SD-WAN using a VPN. A software-defined backbone connects those PoPs.

    By addressing the security and connectivity issues of networking, cloud-based SD-WANs address a larger problem set than software-defined backbones. This cloud-based SD-WAN replaces the functionality of traditional SD-WAN and security appliances including next-generation firewalls (NGFWs), and Intrusion Prevention Systems (IPSes).  

    Since software-defined backbones underlie cloud-based SD-WANs, they too reduce costs. There are, however, significant implementation differences between the two IP-centric approaches. In the case of software-defined backbones, one provider operates over one IP network and touts its optimized routing protocol, whereas the cloud-based SD-WAN constructs an overlay across multiple tier-one carriers, selecting the optimum carrier in real time.

Global WANs beyond managed MPLS services

The days where global WANs depended on carriers and their managed MPLS services have long passed. SD-Core technologies provide enterprises with alternative approaches that allow organizations to reduce their bandwidth spending without compromising network performance. Before you consider SD-WAN, don’t neglect the underlying technologies that can better ensure SD-WAN performance.

For more Cisco news:

For more Cisco products and services:

Steve Garson

Steve Garson is the founder and CEO of SD-WAN Experts and a regular contributor to technology publications, including Network World.