Software Defined-WAN (SD-WAN) Crystal Ball – The Journey to Mainstream
Gartner noted in their recent report “Hype Cycle for Enterprise Networking and Communications, 2017” that the SD-WAN Service had reached the “Peak of Inflated Expectations” phase in their Hype Cycle model. While this sounds encouraging at a quick glance, what this suggests is that the market’s expectations will diminish over time as the real challenges of the technology are revealed (“Trough of Disillusionment”) until the expectations recover as organizations gain true understanding of the technologies benefits and risks, and the technology further matures (“Slope of Enlightenment”). So, what are the challenges and limitations of SD-WAN and what advancements can we anticipate in the future?
In this post I would like to highlight two notable topics:
1. SD-WAN is not a silver bullet that can adjust and fix issues on its own. Due to the way that SD-WAN is generally marketed it is easy to understand this misperception. Organizations should understand that SD-WAN can dynamically steer traffic depending on real-time performance of the circuits, it does not however, promise dynamic adjustments of configuration and parameters such as application SLA thresholds. What this means is that network administrators will need to constantly monitor and review performance and tweak parameters to adapt to the changing traffic profile as new applications are deployed.
Organizations need to account for such administrative work when making the transition to SD-WAN, or consider investing in a managed SD-WAN offering from service providers in order to realize the full potential that SD-WAN enables.
When I gaze into my crystal ball I can foresee that SD-WAN will embrace big data and artificial intelligence (AI) – two key enablers of digitalization — and that this will enable dynamic configuration and parameter adjustments. For example big data could enable a predictive analytics engine that supports execution of remediation measures prior to the occurrence of outages. Use of AI has unlimited possibilities – for instance, parameter adjustment and optimization may be automated by the intelligence gained through machine learning in a way that adapts to policies previously deployed by the network administrator. This may realize the ultimate “software-defined” WAN.
2. SD-WAN is not an end-to-end networking solution. Network planners and administrators require end-to-end visibility and control of their enterprise network to effectively carry out their jobs. SD-WAN, as the name suggests, is an evolution of WAN technology and its network virtualization capability is limited to the transport portion. We must acknowledge the fact that SD-WAN offers limited or no alignment to the networking elements outside of WAN such as datacenter [internal] network, office LAN and security.
To overcome this challenge, some SD-WAN vendors are developing a concept of “service chaining” that allows interconnection of SD-WAN with external appliances such as firewalls. This offers improved integration across multiple technology elements. To push this concept further, the industry may take advantage of Network Functions Virtualization (NFV) technology - the virtualization of network functions such as; routing, security, load balancing and WAN acceleration - to converge these functions into a single communication device, in some cases referred to as white-box CPE.
In summary, SD-WAN certainly has challenges and it may not meet the hype today but my crystal gazing suggest that SD-WAN will overcome its known limitations in the not too distant future. What is important is for organizations to comprehend the reality of SD-WAN, despite its hype, and prepare to adopt the technology at the right time. Contact me and I will look into my Crystal Ball to identify if the time is right for your organization to adopt SD-WAN.
 Gartner, “Hype Cycle for Enterprise Networking and Communications, 2017 (ID:G00314958)”, July 2017, by Danellie Young, Neil Rickard, Bjarne Munch
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.