5 Common SD-WAN Challenges and How to Address Them

5 Common SD-WAN Challenges and How to Address Them

Software-defined WAN is a feature-rich technology that consolidates networking, security, reporting and management into a single platform. But potential users should be aware of the inherent SD-WAN challenges and deployment risk factors.

This article explores five challenges IT teams face when evaluating SD-WAN vendors: vendor selection, underlying configuration, cloud connectivity, cost reduction, and management.

[[398228]]

1. Supplier selection

The first challenge associated with SD-WAN is how to choose a vendor. When starting to evaluate SD-WAN, most IT decision makers start by researching leading vendors. This first step is challenging because there are so many SD-WAN products on the market. IT teams can be overwhelmed by these products, each of which seems to offer digital transformation in WAN services.

Over the past few years, the WAN has evolved to include a variety of technologies, such as secure access service edge (SASE), integration with major cloud providers, and many technologies that help increase application performance and diversity. Security, cloud computing, and new capabilities allow users to work securely regardless of where they are, but IT teams need to analyze their user workflows to understand which vendor can best deliver their services.

To address this challenge, IT teams can document the business application flows that exist today and how the enterprise plans to operate over the next decade. When creating this report, the team must focus on how users can be more productive through SD-WAN capabilities.

When making technology decisions, IT teams are often constrained by their existing contracts and network complexities. For example, security vendor contracts often do not end at the same time as WAN services. In this case, teams should consider vendors that offer good integration with third-party security services. In many cases, existing contracts pose a challenge because vendors are combining various functions into one product.

2. Underlying resource configuration

The shift from dedicated WAN technologies, such as MPLS, to the Internet as the default connectivity option for SD-WAN makes sense. Almost all enterprises are adopting a public cloud-first strategy in SaaS, IaaS, and PaaS. The challenge associated with SD-WAN is which underlying service provider is best for the company's location - and whether to use a single IP backbone or adopt a multi-ISP strategy.

For large, global enterprises, a single IP backbone is a better fit because traffic remains within the autonomous system, resulting in more predictable latency and jitter across application performance. In contrast, national networks are less concerned with round-trip latency, so they can consider a multi-ISP strategy based on a single zip code.

However, teams should consider more than just network performance. Traditional MPLS network operations centers are known for their focus and troubleshooting capabilities, providing end-to-end management of the WAN edge and circuits. Often, ISPs are less focused, which means that it is up to the vendor to troubleshoot and manage connectivity issues. Depending on the management choices made by IT teams (DIY SD-WAN vs. managed SD-WAN), it is important for them to understand how potential vendors will provide service-level agreements for monitoring and troubleshooting connectivity.

3. Cloud Connectivity

In almost all SD-WAN vendor selection projects, IT teams will need to connect to AWS, Microsoft Azure, or Google Cloud Platform. In terms of SD-WAN vendors’ ability to access cloud services, they generally fall into the following three categories:

  • Native cloud access is built into the vendor’s architecture and is linked to branch structure sites using a cloud backbone infrastructure.
  • Vendors deliver their SD-WAN appliances into cloud environments through public gateways or private backbone networks.
  • The vendor leaves it up to the customer to deploy the equipment in their local cloud data center.

In the first option, SD-WAN vendors use cloud computing as their global backbone. Deploying a cloud gateway architecture is not yet common, but it is a wise choice because connecting to the local cloud data center is the ultimate destination for user traffic.

The second option provides flexibility regarding vendor go-to-market capabilities, including proprietary backbones or public gateways that can route traffic more efficiently than the Internet.

Finally, the third option provides access to a cloud provider, but in a temporary, simplified architecture.

4. Reduce costs

Cost reduction is one of the main drivers and marketing strategies associated with SD-WAN. However, cost reduction is often not quantifiable and needs to be considered in terms of the overall benefit to the business. For example, SD-WAN combined with SASE can improve work efficiency. While enabling users to access applications efficiently is not a budget item, the overall impact on the business is huge.

Another standard way to reduce costs is to purchase local site-to-site Internet infrastructure from the lowest-cost service provider. When comparing like-for-like prices to MPLS circuits, enterprises can save a lot of money.

Functional consolidation can also save a lot of money, and IT teams will consider vendors that offer built-in SD-WAN, SASE, and cloud provider access to their equipment. Through consolidation, SD-WAN is easier to manage and use, which can have a positive impact on IT departments because they need fewer resources to manage the system.

5. Management

SD-WAN is blurring the lines between DIY, co-managed, and fully managed SD-WAN. Enterprises don’t have to choose their level of management in the traditional static way. If the vendor doesn’t own the full technology stack, additional expertise must be brought in to deliver the system. This is the way WAN products have traditionally been delivered: by piecing together platforms to build functionality. This often results in a slow process because the systems between each technology stack often don’t work well together.

If a vendor owns the full technology stack, it can help customers manage their WAN based on the needs of any given situation. By maintaining control over the technology, the vendor is well positioned to handle any needs across all levels of managed services. If the IT team wants to manage changes, the SD-WAN management interface makes this easy. If the team needs help, the vendor can help them because they understand every aspect of the delivered product.

In contrast, service providers offer their support for SD-WAN by combining the underlying and upper layers with multiple vendor options. The platform is built across multiple vendors, which can result in a less agile and time-consuming process. If an enterprise does not need flexibility across managed services, it can choose a managed service provider.

<<:  my country successfully developed a 62-bit quantum computing prototype "Zu Chongzhi"

>>:  Graph Algorithm Series: Shortest Path in Computational Graph

Recommend

A brief analysis of the application of NAT technology in cloud gateway

Labs Guide With the increase of network applicati...

Forecast of cellular network development trends in 2017

[[183195]] The new year of 2017 will continue the...

Maxthon Hosting: 56 yuan/month Hong Kong CN2-2GB/40G SSD/300GB/15M Bandwidth

Aoyo Zhuji is one of the foreign hosting services...

"If you don't believe it, run a test?" Is it a gimmick or real strength?

1. Background: The Battle of Performance "If...

The 10 most representative software-defined network tools in 2020

The global software-defined networking market has...

Tested in 6 cities! How fast can 5G uplink with Super Uplink run?

Ever since 5G has entered the homes of ordinary p...