The new market category of Virtual Quality Networks (VQNs) opens up a race to deliver engineered experiences and predictable performance for the cloud. The loser may be well the present Internet, which has an unsustainable technical and economic model.
Every so often, a new sector appears in the telecoms industry. We’ve seen SDN, NFV and SD-WAN arrive recently, as network resources are increasingly placed under software control. This is happening both inside the network operator (with SDN and NFV), as well as at the edge of the network (with SD-WAN).
I believe we are seeing an important and new sector emerge, the Virtual Quality Network (VQN). Just as a virtual private network (VPN) isolates users and applications in terms of security, a virtual quality network isolates them in terms of performance.
Enterprise users are increasingly finding the Internet is a disappointment for their application needs. After all, enterprises greatly value predictability for stable business operations. The Internet is not presently designed to deliver any kind of assured access services. The backbones select routes in order to minimise cost, not maximise quality.
The end result is highly variable quality and unpredictable application performance. This is of particular concern to CIOs who are looking to adopt thin client and SaaS applications. These are crucial to their ambitions to reduce cost, deliver better on operational promises, and enable innovative new ways of working.
However, the risks of “getting cloudy” keep CIOs awake at night. They cannot know in advance if the Internet will consistently deliver on the business process promises they make. The only way to know if cloud will deliver is to do it, which is a dangerous business gamble.
There is no “safety case” for the transition, other than empty vendor promises and hearsay. There has been no magic “turbo” button to press to get out of trouble when they run out of performance slack.
The only current alternative to cheap and unpredictable broadband Internet access is to buy dedicated circuits (with MPLS or Ethernet). This is very expensive, can take a long time to provision, may not always be available, requires long contract-lock in, and involves forgoing a lot of the efficiency benefits of packet data. In other words, it is the antithesis of the “on demand” promise of cloud and affordability of statistical resource sharing.
Whilst the access network may be the biggest QoE bottleneck, it is not the only one. The new VQN service providers create a third option for CIOs, other than “poor access quality” and “high access cost”. This “turbo button” for the Internet is a little bit like how a toll road expressway bypasses traffic congestion in towns and cities. VQNs bleed enterprise traffic off from the “best effort” Internet, and instead deliver it over a managed overlay network.
The VQN overlay network offers managed longhaul connectivity from local points of presence. This connectivity can be contracted from cloud vendors like IBM, Microsoft and Amazon. They have built their own global cloud distribution networks, and they tend to be relatively lightly used (compared to their vast capacity). The VQN can also be constructed by dynamically arbitraging Internet backbone links, depending on their performance “weather”.
In either case, the result is tangibly better than the standard public Internet. It offers the CIO a means to overcome many performance issues by simply turning on the VQN to bypass QoE problems. Even if the access network remains imperfect, the end result is far more acceptable to users. For some applications, like media companies uploading large video files, the difference is a spectacular order of magnitude in performance.
You can think of this as being a bit like how a CDN operates, but instead of being dedicated to serving frequently-accessed recorded content, they are optimised instead for interactive SaaS applications and very large and time-sensitive file transfers. There are already two VQN vendors duelling it out, Teridion (a US/Israeli outfit) and LiveQoS in Ottawa, Canada with their LiveONE service. [Disclosure: LiveQoS is a consulting client.]
You can also consider these VQNs as being complementary to other players who are rethinking how the Internet works, such as 128 Technology, or those in R&D working on RINA. There is a growing acceptance that the first-generation Internet isn’t up to the job that society is asking it to do. It needs refactoring and redesigning, and enterprise users are taking the lead in demanding fitness-for-purpose.
As “premium” enterprise traffic is diverted off the present Internet, it will inevitably have an impact on the economics of what is left. Whilst “net neutrality” activists have focused on the presence of “fast lanes” and “zero rating” in the broadband access network, they have totally ignored the potential for VQNs to bypass the core. This opens up new regulatory issues on how to protect everyday Internet users from plummeting quality.
The current quality rationing model of the Internet way well be facing collapse, as a new “pay for performance” one takes over. Companies like Google are designing and deploying more aggressive protocols to seize network resources in competition with other users. They are effectively engaging in an unlimited printing of ration coupons, since adding demand is free.
When this quality collapse bites, as it already has in places, then it will also have a ripple effect in the rest of the IT market. Device vendors and application developers who have become dependent on “best effort” (from under-priced telecoms assets) may be in for a nasty wake-up call. Technologies like SD-WAN are feeding off a dying broadband service model, and any arbitrage extracted at scale must eventually disappear.
IT and application vendors that decide to opt for control over the network performance, paying for better quality with a VQN, may be making a very wise investment in their strategic future. VQNs signal the beginning of a new wave of engineered experiences for the cloud.
These are delivered by dynamically matching supply to demand for quantity and quality. New business models will bundle together durable hardware with ephemeral services. VQNs open up the possibility of radical business model innovation for SaaS and thin clients.
VQNs also naturally move us towards rational network economics with a proper market pricing for performance, hence are sustainable. Low latency and immediate access to transmission resources ought to cost more in any sane telecoms industry. We must pass money along to reward those who invest capital and create higher quality services.
That telecoms is only belatedly discovering this is very instructive: broadband and the Internet are still immature, and continue to develop rapidly, both in terms of delivery technology and business model. As the present Internet is left with only price-sensitive users engaged in mutually-assured performance destruction, its long-term viability comes into question.
My guesstimate is that VQNs will ultimately de facto become the new Internet. This is especially likely if they also begin to include “fast lane” assured local access to end users, as that much-needed model gains traction. The technical capability exists today, and its spread is inevitable over the next decade as the products and market mature.
The end game is all data being transported in virtualised network containers, just with different quality contracts and pricing models. This is a bit like going from the present “MS-DOS level” Internet to a Windows or Unix one. The security and performance isolation of different users and applications in a better “multi-tasking” Internet is unavoidable.
The dynamic matching of supply to demand also naturally invokes multi-sided market business models, with powerful increasing returns to scale (like eBay or Uber). That means the rewards for early market entry and success are very high, and there may be no “fast followers” in the VQN business.
I suggest that VQNs are a new high-growth enterprise telecoms market to watch very closely indeed.
<h4><strong>For the latest fresh thinking on telecommunications, please sign up for the free Geddes newsletter.</strong></h4>
<div id=”mc_embed_signup”><form id=”mc-embedded-subscribe-form” class=”validate” action=”http://martingeddes.us1.list-manage.com/subscribe/post?u=f105fd56904428bca9da44a82&id=4464a0ef1d” method=”post” name=”mc-embedded-subscribe-form” novalidate=”” target=”_blank”>
<div class=”mc-field-group”><input id=”mc-embedded-subscribe” class=”subscribe” name=”subscribe” type=”submit” value=”Subscribe” /><input id=”mce-EMAIL” class=”required email” name=”EMAIL” type=”email” value=”Enter email address” /></div>
<div id=”mce-responses” class=”clear”></div>
</form></div>
<!–End mc_embed_signup–>
Read It Now
