Is SD-WAN a Quasi-QoS overlay for enterprise, independent of telcos & NFV?

In the last two weeks I’ve been at two events: EnterpriseConnect in Orlando (EC16), and NetEvents in Rome. The former is a midsize trade-show, mostly UC/cloud-comms providers and vendors pitching to business users. The latter is smaller: vendors and a few SPs briefing and debating in front of technology journalists and analysts, mostly about enterprise networks, or the carrier networks needed to support them.

An interesting divide is emerging. Both events involved a huge focus on cloud – especially for communications apps and security functions. But it is mostly only the “traditional” carriers and their major vendors which are really discussing “proper” NFV and SDN as a platform for delivering new customer-facing services to businesses. For other enterprise vendors and service providers, NFV is not even on the radar screen as an acronym.

EC16 was dominated by major players in telephony and collaboration – vendors like Cisco, Avaya and Microsoft talking about cloud-based evolutions of their UC and conferencing tools; UCaaS providers like 8x8 and RingCentral with their own hosted platforms, or others based on BroadSoft. WebRTC, contact centres and cPaaS made a good showing as well. A few traditional telcos were there too, such as Verizon which has an LTE-based UC solution, and Sprint talking about its partnership with DialPad (formerly Switch.co). Slack wasn’t there, but other workstream-style messaging and collaboration tools were pretty ubiquitous, usually with a heavy mobile bias.

There was also a decent turnout of comm-centric vendors that make SBCs, UC/telephony servers and related infrastructure elements – Oracle, Dialogic, Metaswitch, Sonus, BroadSoft and peers. But while these were definitely talking about virtualisation, it was mostly not in the guise of NFV as perceived by the telecoms industry. There wasn’t much discussion of MANO and service-chaining, unless I specifically asked about them in meetings. Their use-cases for virtualisation were all much more pragmatic, aimed at non-telco UCaaS providers, or in-house deployments by enterprises in private-cloud or hybrid cloud/on-premise configurations.

The general assumption was that enterprises will continue to buy their collaboration apps/services separately to buying their network connectivity. Even where a UCaaS provider also sells access or SIP trunking, they’re not likely to be tightly coupled. There might be some "dimensioning" to ensure sufficiently-reliable performance, a separate MPLS connection entirely, or some tweaking of prioritisation to the UC provider's cloud. But there was no sense that a customer-facing UC server would be “just another VNF” hosted in the telco’s infrastructure alongside its vIMS and vEPC.

In a nutshell – corporate telephony and collaboration and contact centres are not really seen as “network functions”, any more than SAP or Office or other line-of-business apps are. (It’s worth noting that security functions like VPNs and firewalls are more aligned with NFV, as they are often integral with access connectivity). There's no real "telco cloud" either, except as an equivalent of an ISP-cloud or SaaS-cloud.

Maybe this will change in future as we see more telco "distributed cloud" and "fog computing" architectures emerge - for example, the Mobile Edge Computing initiative. But to be honest, I've got my doubts about that as well - a topic for another post, though.

However another form of software-based infrastructure for enterprise got airtime at both events: SD-WAN (software-defined WAN). I am starting to think that SD-WAN may actually reduce the potential for some proposed NFV business models, because it could put a new layer of abstraction between telco networks and corporate applications and communications.

In essence, SD-WAN allows the creation of “Quasi-QoS” by various methods. Perhaps the most important is the blending together of multiple access connections to a company’s sites – MPLS, vanilla Internet (perhaps x2 or x3), LTE and so on – and then load-balancing, bonding or using them for backup or differential routing of traffic.There are also various approaches involving hacking TCP in some way, or various proprietary approaches to packet classification and scheduling. Typically, SD-WAN will involve some sort of server or dedicated box at each customer site.

The following illustration is from SD-WAN vendor Velocloud, and is given as an example

This could put mission-critical applications onto managed connections, with less-demanding traffic onto Internet access. Or it could be “Internet-primary” and use more expensive connections only when congestion seems to be causing problems. It can also link into major IaaS and cloud platforms (Amazon, Google, Microsoft etc) in different locations and with large-scale connections. Many other use-cases and permutations are feasible as well, especially when linked with UCaaS or other SaaS offers.
 

In other words, SD-WAN could be described as “Arbitrage-as-a-Service” or “Managed Least-Cost-Routing”. Where the SD-WAN is offered by a company which isn’t one of the access providers, it is essentially “OTT-QoS” – although I think that “Quasi-QoS” sounds better.

I see this as a conspicuous threat to various forms of NFV-based enterprise service, especially what gets called NFVaaS or NaaS. By putting an overlay around access connections, it reduces the ability for any extra capabilities to be offered from within their infrastructure.

My colleague Martin Geddes has been scathing about this type of “QoS” (link) – noting that the underlying “network science” doesn’t allow for performance to be accurately predicted or guaranteed, without some very clever maths in the access network boxes. The “failure modes” can be ugly, as sudden sub-second spikes and buffering issues can occur and disappear randomly, trashing sensitive applications before the SD-WAN can respond.

My sense is that while that might be technically true, the real-world problems are more prosaic. Either a fully-dimensioned MPLS connection is too costly, or something fails completely because a fibre is cut, or a network node crashes and reboots. Or, is is the case here, the economics are so compelling that it's cheaper to just buy two redundant connections rather than optimise one.

The bottom line is that SD-WAN is potentially a game-changer - and it potentially undermines the NFV argument, not just for UC services, but perhaps other functions too. While some vendors are working with telcos to offer hybrid solutions, that's because of customer pull. This isn't to say that it invalidates everything proposed by NFV believers - far from it, in fact - but it does act as a counterbalance to the view that virtualisation is all telcos need to dominate enterprise connectivity and communications. 

SD-WAN entrenches the idea that enterprise communications and apps are decoupled from access. It also empowers Internet-based UCaaS providers to offer SLAs and QoS guarantees without owning access connectivity themselves - for example, Vonage works with VeloCloud, and Star2Star has its own connectivity boxes that optimise "vanilla Internet" access.

Videoconferencing does not replace business travel

I did a short radio interview the other day, about London airport expansion, and the possible new runway at Heathrow or Gatwick. I wasn’t speaking about the choice of one location vs. the other. Instead, I was speaking as a frequent traveller, and why additional capacity is likely to be needed in the first place.

Some of the reasons are aviation-related: congested airports have no wiggle-room if something goes wrong. I’ve suffered big delays when weather or other incidents has meant that airports can only operate at reduced capacity, with greater spacing between landings and take-offs. Then there’s a desire for more direct flights from London to cities across Asia, the Americas, and MEA. Changing planes in NY or Singapore or Istanbul is always possible, but that just adds more time (and additional fuel-heavy take-offs). Various other reasons apply too.

But I’m not an aviation specialist. I’m a communications industry analyst & consultant.

What really stumped the radio-host was when he suggested videoconferencing might replace most business travel, so fewer flights would be needed and therefore perhaps less airport capacity.

I responded that ironically, about half my own travel is to events/clients actually involved in the videoconferencing industry, or in other aspects of advanced communications. I regularly attend and speak at video, UC & WebRTC events in person, have private workshops with operators or vendors, meetings with investors and so on.

Could I do some or all of these via a phone-call or video session? In theory yes, some could be done remotely, but in my view they would be much less productive – and many wouldn’t happen at all. This isn’t just my personal dislike for video either (I prefer voice-only, in general), but a more general observation.

Thinking about it, I’m pretty sure that the people who pitch online alternatives as a replacement for in-person meetings probably don’t do much of either. There are at least 10 reasons why audio & video-conferencing is not a replacement for business travel.

Firstly, it is worth noting that the applicability of video/collaboration tools will necessarily depend on the type of meeting involved. There are multiple “use-cases” for physical business travel, each with different characteristics:

• Short one-to-one meetings (maybe 1hr in length) for sales calls, introductions, catch-ups with colleagues etc. These can be sub-divided into company internal meetings (eg boss/employee) or external (eg salesman/client) which have different dynamics
• Internal small-group meetings, eg a project team distributed across multiple locations. Again, these can be internal or external (eg consultant presenting to the board)
• Site visits, where someone is shown around multiple physical parts of a location, has a variety of meetings & so on.
• Trade shows where the emphasis is on booths and the “show floor”
• Conferences of 1-3 days duration, with multiple presentations, panels, break-out sessions etc
• Seminars (maybe 1-3hrs) with a few speakers and predominantly “broadcast” mode with some Q&A
• Interactive workshops where people interact in small groups
• Team-building sessions combining a mix of presentation and social/bonding activities
• Many other types of “meeting”.

While any of these can use the same transportation mode (ie a flight) they would all need to be re-invented with different forms of conferencing or collaboration application. Some are easier than others – informal meetings with a small dispersed project team, for example, can be done with a simple audio or video bridge, ideally with file/screen-sharing as well. Webinars can replace some seminars.

But a full-on trade show, with demos and new products, as well as private meeting spaces for confidential discussions, cannot really be replicated online to any reasonable extent. Neither can good interactive workshops, or even summit-type conferences. As a regular panel moderator and conference chair, I don’t think anywhere near the same experience could be done via video as in person. Maybe in 10 years time, with Oculus Rift version 8 and some advanced haptic interfaces and full body-suits, but I’m not convinced.

Some of the limitations of videoconferencing-style replacement for physical meetings:

• Lack of detail – while you can replicate lifelike scenes with 4K video, it’s still not fully immersive without stereoscopic vision, ultra-fast frame rate etc.
• No way to support culturally-important actions like handshakes or physical exchanges of business cards
• Security and privacy – how can you be sure that the quiet chat over a virtual coffee remains confidential?
• Subconscious awareness of body language and micro-expressions
• Cognitive absorption – what part of your concentration is diverted to seeing how you appear on-screen to other people?
• Technical complexities of managing virtual events with multiple parties, using different networks & devices. WebRTC and its peers only go so far
• Dependency on camera/sound crews, cameras, microphones – which then mean you get an “edited” version of an event rather than your own choice of where to sit/stand/walk around
• Lack of sync between timezones. Do you want to get “virtual jetlag” by attending the breakfast session at 9pm at night in your timezone & listening to conference presentations until 4am?
• How do you facilitate networking over meals, provide “back-channels” to whisper to your neighbour during sessions, manage realistic arguments or back-and-forth discussions and so on?

Overall, while online collaboration is OK for some use-cases, it is generally a second-class citizen, with numerous almost-intractable limitations. It would reduce the effectiveness of companies, compromise security and productivity, and advantage people with geographic proximity.

In many ways, videoconferencing is becoming more important. In future we may have access to contextual communications tools which may improve some interactions so they're better than real-life speech and vision. But it’s usually more accurate to consider it as a “better phone call” or “richer than an email exchange”. It’s a big step down from interaction in person. Conferencing can enable extra conversations, or allow extra people to attend existing physical meetings remotely. But that is not the same as replacing the core in-person conversations.

Ironically, better remote conversations may lead to more international business and travel. A more-effective initial introduction via video/voice may well lead to new relationships being built. And later, those relationships will often involve in-person meetings, for site visits, events, interactive workshops and so on. Certainly, without my extra “reach” via both conferencing and social-media, I wouldn’t have nearly as many international clients to work with.

It’s also worth noting that while videoconferencing might be able to replicate some aspects of traditional meetings, the latter have evolved as well. Many conferences now employ techniques that are experiential or immersive. Group exercises, interaction with voting terminals, not to mention the improved venues and carefully-crafted social interaction episodes.

A similar story is true for consumers. Wearing a virtual-reality headset in a tanning salon is not a substitute for feeling the beach sand between your toes. Videoconferencing into your distant family’s Xmas dinner doesn’t work, if you can’t taste the turkey and pull the crackers. Listening to a rock concert on the radio doesn’t compare to jostling and jumping with other 

fans of the band in the arena.

Nobody can conclude a deal with a video-handshake in a virtual restaurant, or experience Burning Man by conference-call. 

Those are "contexts" that cannot be replicated online.

You have to be there.

Initial thoughts on Enterprise Connect #EC15 & WebRTC

I'm in Orlando at the big Enterprise Connect conference and trade-show, covering business communications, especially for large corporations and government. Its main focus is on unified/disunified communications, collaboration, and contact centres / CRM. Yesterday was the WebRTC "conference-in-a-conference" and today is the main session with kickoff keynotes (and various briefings), although yesterday had some general plenary panels too.

Note: this was written on March 17th 2014, so "yesterday" refers to the 16th While I'm here mostly as I was invited to speak on a WebRTC panel, it's interesting for me to compare and contrast EC with many of the more telecom operator-centric events that I go to, on a more general level.

The first thing is that there seems to be much more pragmatism and individuality among enterprises than among telcos - they are focused on solving their specific problems, in their industries and specific geographic regions. There is much less need for businesses to feel the need for "industry-wide" solutions and standards, and more willingness to experiment and customise.There is a sense that most big companies will have their own in-house software development teams, as well as relying on systems integrators, consultants and a variety of vendors, cloud platforms and other systems. Few businesses have cookie-cutter communications functions or apps, even if the ultimate technology building blocks (eg SIP, HTTP) are the same.

The second thing is that there is much more of a sense that the "end-user" - whether an employee, an external partner, or a customer, is in a position of choice and control. While there is a slight wistful feeling that everything used to be much simpler when there was just a phone, fax and email, there is also a sense of excitement that new developments lead to improved productivity and better engagement. 

There's no disparaging reference to "OTTs" (or spiteful over-defensive recourse to regulators), merely a pressure on suppliers to come up with better, cooler communications systems that fit employees' or customers' needs. While security and compliance are important for many, so is customisation and empowerment. I hear phrases like "use the best tool for the job" a lot, or "multi-channel", as well as BYOD, cloud services and "customer journey". There's a recognition that employees or customers will migrate to whatever application or interface mechanism best suits them - and it's up to the enterprise and its suppliers to get it right, or give developers the right APIs so they can instead.

The third thing is that enterprise communications vendors - some of them the same companies that also sell to telcos - have a much greater awareness of user-interface design, the need to fit their products around users' purpose and workflow, web/app integration, and the need for optimised functionality. Cisco, Microsoft, Avaya, Google, Unify and a whole host of others are at pains to show how slick their mobile apps are, or their ability to allow business processes to be adapted with visual tools. The word "workflow" is uttered 10x more than I'd ever hear in a telecom context.

Other things:

• I see the word "context" everywhere too. This is definitely the epicentre of the Contextual Communications trend I've been mentioning - and I can see it evolving further with the use of WebRTC, sensors, big data analytics and so on. Compared to the "any colour you like as long as it's a phone call" mentality of telecoms, it's refreshing. So much for VoLTE being "innovative". Aspect and Altocloud are among those doing context-based WebRTC applications for contact centres.
• Cloud is everywhere. That's not really news, but it is striking.
• Microsoft Lync (now clunkily renamed Skype for Business) is everywhere too. How it supports ORTC/WebRTC will be critical in future - although as every competitor has a full API of some sort, I'd be surprised if S4B doesn't as well.
• Messaging and timelines are everywhere. All the new collaboration tools look like they're heavily influenced by Facebook, Twitter and the like. This is good. 
• Apart from Mitel (which just acquired Mavenir), nobody seems bothered by the idea of integrating with telco applications, VoLTE or (I'm joking here) RCS. I asked a major telco's business videoconferencing unit representative if they'd be using ViLTE/IR.94 in future, and he didn't appear to have heard the term before.

WebRTC is also heavily represented, but often as a means to an end: an enabler or option, rather than a holy grail of some sort. That said, it's telling that the keynotes all now assume that everyone knows what WebRTC is, rather than introducing it as "a new technology for putting comms into your browser" as is still common at other events. It's also clear that WebRTC is not just "production-ready" but is also in polished, real products being bought and used in anger.

Quite a lot of new products are noticeably WebRTC-based, such as collaboration solutions like Unify Circuit and Cisco Spark (the renamed Project Squared) and the OnAvaya cloud contact-centre solution hosted on the Google Cloud (and with Chromebook integration).  That said, most of these also have other non-WebRTC options, either for legacy browsers, or where a better set of functions can be delivered via native applications. 

The Avaya/Google collaboration potentially points to really rich contextual customer-service features in future, depending on how much analytical and insight horsepower Google can bring to bear. Think about routing to different agents, using different scripts, plus video and other features, where the caller is using Android or logged into their Google account.
 
Various WebRTC-based PaaS and SDK providers are present here as well - Twilio, CafeX, GenBand Kandy have booths, and I've bumped into people from Temasys and Respoke (Digium) as well, while Avaya has an SDK to video-enable customer-service apps too. Interestingly, Vidyo is touting its own video-enablement  SDK, which is not WebRTC-based, although it is working with Google to get SVC capability working on the VP9 codec.

There's also a lot of interest specifically in mobile apps with embedded video-calling for customer service. Avaya demo'd a golf-related commerce app, which it said it said it had added video to in just 2 days with a single engineer. Other vendors also touted their ware for app-embedded video, and I was taken out for an evening by a major US enterprise, to discuss how they could enhance their customers' experience with something similar.

But perhaps the most eye-opening bit was the presentation by a number of enterprises - Medweb, UTHealth [University of Texas] and American Express - about existing, long-standing WebRTC deployments and their learnings from their use and deployment. Some are using it to improve vanilla videoconferencing, but others are integrating it into apps and workflows much more extensively. Medweb has a telemedicine kit which can give a video consultation while streaming output from a medical device (eg a USB-plugged ultrasound scanner) over WebRTC datachannel.

The bottom line: enterprise retains its lead in WebRTC in terms of sophistication of use-cases. While there's various cool WebRTC consumer apps - as well as big guns like Snapchat - it's really the corporate uses that are the state of the art. Contact centre use of WebRTC is nothing new, but it's definitely the turn of UC/collaboration to take the stage at the moment, with app-integration and workflows the next in line.

In fact, I'm starting to suspect that one of the main near-term opportunities for telcos with WebRTC is within their internal IT and communications infrastructure, rather than new subscriber services. At MWC there was an announcement of a WebRTC-powered video contact centre for operators, and at EC the lead customer on stage for Cisco Spark collaboration was from Telstra.

Note: Disruptive Analysis' research and forecasts on WebRTC include detailed coverage of enterprise use-cases, as well as telecoms and consumer web. For more details on the report click here or message me if you are interested in private consultations and advisory services.

Dis-Unified Communications: WebRTC & The Enterprise App Dilemma

Enterprise technology often follows consumer trends. Currently, in the mainstream telecoms world, phone call and SMS usage is being eroded by competing and fragmented IP and app-based platforms – the so-called “OTT players”. Instead of using a handful of standardised (but expensive and limited) services, people are voting with their thumbs and using a wide range of communications apps and tools, on a case-by-case basis. They offer a mix of functionality, low/free cost and "coolness" - Whatsapp, Skype, Kik, LINE, WeChat, Viber, Hangouts, iMessage, Instagram, SnapChat and so on.

In business, an equivalent trend (or threat) is brewing. It manifests as communications dis-unifying, despite the industry buzzword suggesting "unified" platforms are the only path to the future. As well as standalone VoIP and conferencing systems, we are also seeing specific instances of enterprise communications being carved out of traditional, centralised UC/PBX platforms and integrated directly into cloud-based applications and business processes.

In addition, business users who have never before had a PBX "seat" or extension (eg the self-employed, or startups) are using sophisticated mobile or cloud-based apps to enhance their communications experience. They are also using IP-connected phones and tablets, rather than legacy circuit-based communications.

Some dis-unification has always been inevitable - we all get invited to other companies' conferencing systems, webinars or even messaging platforms. Many in the business community use their own personal mobile and home phones, Skype when travelling, send messages with personal webmail - or indeed using the tools built-into sites like LinkedIn and Twitter. New collaboration tools like Slack are currently getting a lot of attention as well. These tools have mostly been used for communication between companies, but a fair amount (especially BYOD calls and SMS) has been between colleagues internally as well.

The question is what happens in future to business communications.

Disruptive Analysis sees two possible paths:

• Towards more fully-Unified Communications, where enterprise IT departments, working with their UC&C vendors and service providers, extend corporate communications platforms to mobile, to collaboration both internally and externally, and (via APIs and SDKs) blended into corporate line-of-business apps and web tools.
• Towards ever-more fragmented Dis-Unified Communications (DUC), where a declining fraction of a business's total voice, video or messaging traffic transits its own "platform", as employees extend the BYO model from devices, to BYOconferencing, BYOcontactcentre, and adopt applications and cloud-services that have their own contextual comms built in by their developers or service providers.

Disruptive Analysis believes that both trends will happen simultaneously, in fact - but just in different contexts. And WebRTC will be a catalyst of both - allowing UC to address more devices and use-cases, but also allowing "DUC" to encroach from adjacent software domains. (Non-WebRTC approaches will also continue as well, but most will converge over time, such as Microsoft's Lync Web API and ORTC)

We already see UC systems being extended outwards - for example "guest access" enabled in Cisco Jabber, or federation from MS Lync. WebRTC will extend this trend further - already seen in early implementations like Unify's Circuit. Hosted UC platforms offered by telcos are also starting to embrace WebRTC - a trend that is likely to be embraced by most vendors and major operators during 2015.

 

As in consumer markets, a core part of WebRTC-extended business UC proposition is around interoperability and identities being “anchored” in central systems, with well-defined numbers and identity (and security). This certainly makes sense for industries and companies with strict compliance and call-recording needs, for example - and it seems likely that these capabilities will be integrated into line-of-business apps such as doctors' practice-management or banks' trading platforms.

Conversely, other contexts have counter-arguments that communications-as-a-feature works better as optimised “islands”, disconnected from traditional UC platforms. A putative video-interviewing service for recruitment, built into a site like LinkedIn, would probably have no relationship with E.164 phone numbers or single-enterprise IDs. Similarly, clunky phone-number dial-ins for conference calls and webinars will gradually fade out - probably replaced with email addresses or other log-on IDs.

The real battleground, however, will be future vertical applications for business, both internally-focused and for B2C/B2B interaction. There seems little reason for an oil-company's field maintenance application (which might incorporate video interaction for remote diagnosis of problems) to interface with the UC system used at head office - especially if it is cloud-based and used for multi-way conferences with partners and suppliers as well. A security application, which includes both physical guards as well as CCTV cameras, would also likely be standalone.

The choice of communications components would up to the app developers, who may well choose a WebRTC platform provider, and look to support a broad range of endpoints, both with and without telephony capabilities. B2C interaction platforms on the web may also be decoupled from phone-based contact centres - although some will undoubtedly be multi-channel . In many instances, new models of communication will be invented - a permanent telepresence "window" between two offices doesn't really fit with the "call" and number-based UC system. The benefits of dis-unification (to users and developers) would likely outweigh the management/organisational downsides of UC.

Some of the DUC applications will be standalone communications services (as Skype, Talko and many conferencing/webinar platforms are today), while others will integrate audio, video or IM functions directly into other software or websites. Many, but not all, will based on WebRTC, perhaps using developer-centric 3rd-party platforms like Genband Kandy or Acision Forge.

The ramifications of this trend are huge - and largely under-appreciated by the UC community. Globally, there be 1.7bn people using communications as part of their work by 2019 - but Disruptive Analysis forecasts only 23% (400m) will have a formal UC/IP-PBX "seat" - and even most of those will use some form of DUC applications at least occasionally, as well as their main "corporate" system.

The other 1.3bn will use a mix of standard telephony (mostly mobile, GSM/UMTS plus a small % of VoLTE), or various form of VoIP, Video and cloud communications. Even among the hard-core of global "knowledge workers", there will be 600m who use WebRTC at least sometimes in their job - yet fewer than 100m of those will be using it in the context of their UC system. The vast bulk of business interactions via WebRTC will be dis-unified.

Put another way, by 2019, the number of dis-unified communications users in business will probably be twice the number using UC, whether premise-based or hosted. This is as big a trend in enterprise communications as so-called "OTT" apps are for telcos. 

For a full analysis of the impact of WebRTC on UC & DUC, see Disruptive Analysis' research report, which covers the enterprise market as well as telecoms and the consumer web. Purchasers will also receive an upcoming update document. Please click here for details.

Dean Bubley, founder of Disruptive Analysis and author of this article & the report, will be at Enterprise Connect in Orlando on March 16-17, and UCExpo in London on April 21-22. Please contact information AT disruptive-analysis dot com for briefings, speaking engagements and private workshops/consulting