Networking / Technology / Contributed

Extend Your Organization’s Reach with Software-Defined Wide Area Networking

4 Nov 2019 12:03pm, by

Ashwath Nagaraj
Ashwath Nagaraj is co-founder and CTO of Aryaka, responsible for building the company’s vision and technology since its inception in 2009. He previously founded Allegro Systems; a security startup acquired by Cisco in 2001. Ashwath was also a founder of Assured Access Technologies, which developed WAN access and aggregation products. AAT was acquired by Alcatel in 1999. He holds 17 patents in storage, security, architecture and networking.

SD-WAN (software-defined wide area network) at first glance appears to be an almost impossible chimera of “cheaper, faster and better” at connecting branch locations to each other and the central hub or headquarters. By connecting multiple branches back to the center via low-cost services and utilizing centralized, policy-driven management for dataflows, an SD-WAN will fundamentally change the way your organization operates. Building a business case to adopt SD-WAN requires taking into account the savings realizable not only in costs, but benefits in terms of agility, flexibility and simplicity. Deploying SD-WAN can cut WAN service bills drastically and what it offers in guaranteed failovers, almost non-existent downtimes and sheer ease of installation — in days, not weeks — would make any C-suite executive prick up their ears.

Take a Step Back – What Is SD-WAN?

SD-WAN networks all your branches, data centers and cloud computing together into a mesh system that turns all locations into one single pool accessible to all services. By distributing computing power and access across the entire network, instead of specifying it location-by-location, the SD-WAN infrastructure makes the entire network more efficient and faster. Overlaying logical WANs enables an organization to reduce lag time and define needs based on location.

For example, a bank could bring a mobile ATM truck to a music festival or an area impacted by a natural disaster, in order to provide attendees or residents with easy access to cash they otherwise would not have had. This “branch” would not require the same access to services and software that tellers at the bank’s HQ or loan origination desks would need, so a software-defined WAN would be able to allocate data resources to only the services needed — cash disbursement and general ledger access—and deprioritize unnecessary services like loans.

Policies for each machine, whether a PC, a router, optimizer or any other networked device, are centralized. Behaviors for each can be defined and implemented across the entire network without requiring machine-by-machine configuration. And, with fully-managed SD-WAN, this can be done remotely by the service provider, delivering even greater savings in labor costs.

SD-WAN deployments fall into two categories — overlay and in-net. Overlay SD-WAN devices can (in some/most cases) replace entire network stacks, either behind routers or replacing them as the branch WAN connection. Because this becomes a layer of enterprise infrastructure, it is easy to swap out network service providers and link types (T1 versus fiber, etc.). However, this layering also introduces another layer of management and potential challenges with visibility and control. In-net SD-WAN is also known as managed SD-WAN. In this case, a branch location would connect to the nearest Point of Presence (PoP) and the data is served up by the provider’s infrastructure.

Why Implement SD-WAN?

Many WAN-enabled organizations use MPLS, but an April 2019 study by IDC found that 78% of organizations have either deployed SD-WAN technology or plan to do so in the next 12 months. This does not necessarily mean MPLS is going the way of the dodo. In fact, it may not be strategically prudent to simply replace an organization’s MPLS infrastructure wholesale. It may be best to gradually substitute in SD-WAN infrastructure over a period of time, as lower-cost connectivity options present themselves, including business internet services or even consumer-grade broadband. One major area of consideration is an organization’s future growth needs. Those looking for greater savings over a long timeframe would be wise to look into SD-WAN, for example planning all-new branches to be integrated with SD-WAN, and then upgrading older branches when necessary.

Because there is not a large amount of infrastructure involved — In many cases, one simple box is all that’s needed, not a full stack of machines — deploying SD-WAN is much faster than legacy WAN solutions. A new branch could be up and running in 24 hours to a week, versus 30-90 days of installation and configuration with traditional approaches. It’s easy to see how every day’s delay in go-live time translates to lost revenue.

In addition to the speed and agility in getting closer to the customer, one of the most appealing aspects of SD-WAN is the centralized control in digital transformation—being able to easily upgrade and incorporate new technologies as they come available. Data flows from branches can easily be crunched and analyzed at the data center or HQ, and real-time information can flow back. For example, a clothing retailer can offer free WiFi in their stores, and when a loyal customer enters and connects, the SD-WAN system can send the mobile device information back to the cloud, where it is correlated with that person’s account and instant discounts or special offers can be sent to that device for in-store use. If an item is not available at that location, the inventory system can be alerted and shipment arranged, all to make the customer’s experience that much greater. In a retail situation such as this, managed SD-WAN can prioritize application bandwidth on a branch-by-branch basis, so a west-coast location may require more mobile-friendly applications while shoppers are still in the store, and the east-coast branches shift resource allocation to the inventory database as they count items after store closing.

What’s in It for Me?

Because of the flexibility of infrastructure required in overlay or in-net SD-WAN, major cost savings are common. In some cases, simply running the locally available commercial internet through an SD-WAN “box” is all that’s needed. In others, a combination of an SLA-managed SD-WAN core with broadband internet access will be more appropriate. The beauty of SD-WAN is that it can use any available network routes, even legacy MPLS, to deliver the promised level of service. Where large, inexpensive internet links are prevalent, traffic can shift there, de-prioritizing expensive MPLS links. This range of connectivity options means cost curves can shift downward over time. The much smaller stack footprint — moving from a server room to a single box — brings capital costs down, as well.

IT departments always seem stretched thin on resources, even when fully staffed. Switching to a fully managed SD-WAN deployment with 24/7 support anywhere in the world has helped some organizations realize a 90 percent time savings on network troubleshooting.

In the end, it’s up to every organization to put serious thought into whether SD-WAN is right for them. But given the ease of installation, the savings and the ability to not only be up and running in record time, but able to give your customers the experience they want today, it would behoove any IT department to look into it.

Feature image via Pixabay.

A newsletter digest of the week’s most important stories & analyses.

View / Add Comments

Please stay on topic and be respectful of others. Review our Terms of Use.