UC Insights Logo
UCI Bannner

Other insights into UC>>

The Implications of UC as a Cloud Service


by Russell Bennett, UC Insights

December, 2010

Leave a comment

Introduction

One of the latest major innovations in internet technology is so-called ‘cloud computing’, or computing services being offered from vendor data-centers on a utility model.   The National Institute of Standards and Technology defines ‘cloud computing’ as follows:

“Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”

The term ‘cloud’ refers to the standard graphical icon of the internet as a physically abstract, location independent service.  This icon was originally borrowed from the depiction of telephony networks, so it is interesting to note that one of the services that is emerging as a cloud computing service is Unified Communications (UC).

I contend that gaining access to UC from the cloud is nothing new: the current enterprise UC technology derived from free consumer services such as the large Instant Messenger (IM) services, Skype, etc., all of which are hosted in remote data centers accessible via the internet.  However, with commercial UC, this notion is being extended from a free ‘best efforts’ social service to one that is more aligned to business communications and is therefore a substitute for traditional voice networks provided by the telephony carriers.

One interesting aspect of the development of UC thus far is the emergence of new vendors in the enterprise communications market sector, i.e. the major enterprise technology firms.  While the emergence of cloud-based UC is arguably not a significant departure from the current model of hosted telephony, the implications for all stakeholders are far reaching.  The issues that the UC cloud vendors must address will test the limits of their ingenuity in the fields of business strategy, governmental liaison and business development as well as technology.

The Benefits of UC as a Cloud-based Service

To fully understand the implications of cloud-based UC, let’s first examine the benefits of UC for various types of customer.  Note that there is an underlying assumption that large enterprises will almost always choose to host their UC service in their own data centers, because:

  • The enterprise data centers themselves are largely a sunk cost so the cost of hosting new functions (e.g. UC) is already partially amortized into the cost of the data center;

  • Business communication is viewed as a mission critical service – so the value of being able to manage one’s own system failure plan and disaster risk management is compelling;

  • The larger the firm, the larger the total volume of communications, however, the ratio of internal to external communications is also larger, therefore cost savings on self-hosted UC increase proportionately.

Advantages for Firms of all Sizes

Quite apart from the compelling value proposition of the UC user experience, the advent of cloud UC has benefits for businesses of all sizes.   Cloud-based UC enables all business related communications, not just those between large enterprises, to benefit from the richer UC user experience.

Communication with consumers

Many businesses have consumers as customers, therefore the imperative of being able to communicate with consumers in richer ways is becoming increasingly self-evident, e.g.:

  • The emergence of social networks (e.g. Facebook and Twitter) as business communications tools;

  • The use of other forms of internet communications such as blogging, instant chat and voice.

While large enterprises have the resources and negotiating power that better enable them to build connections to consumer oriented networks and services, the exponentially larger resources of a cloud vendor amortizes these ‘peering’ costs across the entire customer base.

Communication with the value chain

Business communications that are not with consumers are with the value chain, both upstream and downstream.  Cloud-based UC will enable all parts of the value chain to gain access to UC user experience; thereby exponentially expanding the value of UC communications (see Metcalfe’s Law later in this paper).

Advantages for Small and Medium Sized Firms

Outsourced communications

There can be no doubt that ‘self-hosting’ a unified communications system is a more complex and expensive undertaking than a small business can contemplate.  Small businesses, by definition, do not support the organic support functions that are found in large enterprises; and the smaller the business, the less likely it is that staff are employed to work on anything other than the main function of the business.  Indeed, if a small business currently outsources telephony to a hosted service then it makes perfect sense to outsource unified communications.  However, due to the ‘smallness’ of a small business, there is also a proportionately higher likelihood that any given business associate is going to be outside the company.  Therefore the notion that unified communications facilitates richer communication between virtual teams is a more compelling argument within a smaller business than a larger one.

Access to ‘big company’ UC

For a small business to be able to compete with larger businesses, much less to become a larger business itself, it needs access to the tools that large companies are using.  Furthermore, to the extent that a small business is part of a larger businesses value chain (see above) then the ability to collaborate with customers and suppliers in large businesses is of paramount importance.

Advantages for Large Firms

As has been noted above, the general assumption is that large enterprises will self-host their UC facilities because of the economies of scale they can achieve.  However, as will be discussed later in this paper, centralized facilities do not scale perfectly in a geographically distributed network.  Clearly, a single data center is a bad idea for failover reasons.  However, various attributes of the organizational structure will be a factor in the design and placement of data centers, including:

  • The breadth of geographic distribution – e.g. many global multi-nationals have operations, sales or support in over 100 countries;

  • The degree of centralization – i.e. what proportion of staff work in corporate or regional headquarters vs. those in local branch offices;

  • The amount of local (downstream) business communication being made from branch offices vs. upstream (i.e. to district headquarters).

For enterprises with sparsely distributed networks in some parts of the world and centralized networks in others, if will certainly make sense to consider a hybrid approach, i.e. self-hosting in the centralized regions and cloud-based UC in the distributed regions.  Of course, the virtual network will remain centralized and under central control, however the cost of supporting (for example) a two–person representative office in remote regions can be significantly reduced via the creation of hybrid networks with local vendors.

Advantages for Consumers

As previously stated, consumers have long had access to many of the features of UC in the cloud in the form of the services provided by Skype, AOL, Yahoo!, Windows Live etc.  However, in the context of businesses gaining access to UC in the cloud, the main advantage is for consumers to be able to connect to these businesses via inter-network peering.  Peering between the consumer UC services is not yet a reality, since they jealously guard their ‘namespace’ as a result of their interpretation of Metcalfe’s Law (discussed later in this paper).

Technical Implications

Latency

The main difference between UC and most other computing services that can be placed in ‘the cloud’ is that most UC modalities depend on real-time media.  This is to say that whereas other services, such as email or internet search are asynchronous (i.e. you make a request and wait for the answer), real-time services are synchronous, i.e. the timing of the arrival of media packets is critical to the usability of the service.

As anyone with experience with long-distance telephone calls will know, the user experience can rapidly degenerate from ideal to unusable because of latency, i.e. the delay between one party speaking and the other party hearing the speech.  One way latency of less than 150ms is ideal, but with latency between 300ms to 500ms conversation becomes difficult.   With a latency of over 500 milliseconds, conducting a conversation becomes nearly impossible.

Latency can be caused by several factors:

  • The speed of light – the natural physical limit of electrons or photons (in the case of an optical network) traversing the distance between the parties;

  • Processing – various types of hardware or software between the parties performing operations on the data packets, e.g. encryption/decryption, compression/decompression; encoding/decoding and transcoding (the conversion from one media codec to another);

  • Network congestion – at bottlenecks in the network, the router software will place data packets into a queue (or buffer) while the backlog is being cleared.

    • Note that latency becomes even worse with every additional network junction that the data packet has to traverse, since route processing and possibly buffering will be applied each time.

Cloud-based UC can potentially increase latency by:

  • Sub-optimally siting and distributing data centers, causing media packets to take a circuitous route between the calling parties (where media is required to traverse the data center);

  • Deploying data centers with a sub-optimal balance between economies of scale and the ability to process communications transactions.

As the cloud-based UC business is ramping up, service providers must carefully balance their investments and initial target markets to ensure that early adopters get the best quality of service.  As more businesses convert to UC, that growth must be carefully managed to ensure that quality of service is maintained via the judicious deployment of additional data centers.

Increased Bandwidth

UC modalities have much higher bandwidth requirements: compare 9.6Kbits in each direction for analog telephony vs. 32Kbits for G.729 and 64Kbits for G.711.   For high definition video, the bandwidth requirement can vary from 500Kbits to 1.5Mbits.

The effect of the high bandwidth utilization is that during peak traffic periods a UC cloud data center could be receiving and transmitting truly enormous amounts of data, which will create network congestion.  Additionally, small business networks (i.e. LANs and WANs) are not designed to handle UC media codecs and will need to be upgraded.  There remains an open question around whether service providers are able to carry such high data volumes for thousands of small businesses in aggregate, or what price they would charge for the bandwidth if they are not also the UC service provider.  This raises the issue of net neutrality, which will be addressed later in this paper.

Scaling Issues

In theory, UC technology provides an elegant solution to the issue of scaling to handle a very large volume of high bandwidth communication sessions being routed via centralized, large scale data centers.  The media (i.e. voice, video and data collaboration) packets are routed directly from user to user, without the need to traverse the UC data center, thereby significantly mitigating data bottlenecks at UC data centers.  However, while this holds true of 2 party communications, the media packets of multi-party communications must still be centrally aggregated and this raises the risk of communications latency discussed above.  Note that even a conference call between parties in the same building must still be aggregated or ‘mixed’ in a data center that could be hundreds or thousands of miles away.   Furthermore, there may be regulatory requirements for media to be centrally aggregated: e.g. for call recording and/or lawful intercept – see the Regulatory section later in this paper.

The implication of many thousands of high bandwidth media sessions being centralized and processed at a data center is that the data center facilities must be carefully engineered to deal with peak traffic volumes.  In theory, a UC data center that carries traffic from numerous time zones can smooth out the peaks as the business day travels from east to west across time zones.  However, the requirement to minimize media packet latency conflicts with the goal of smoothing out traffic peaks.  There are several approaches to mitigating this issue:

  • Deploying smaller, more local data centers rather than very large ones: this approach is at odds with the requirement to smooth traffic peaks, but it does mitigate latency issues.

  • Providing local conference/recording servers to UC cloud customers to host on their own premises to reduce local latency and the volume of data hitting the data center: this approach is somewhat at odds with the notion of outsourcing communications; however, if the server is centrally managed then that issue could be overcome.

  • There is also the possibility of getting conference media to flow from each party directly to all parties without the need for a centralized conference server; the ‘media mixing’ would be conducted on each endpoint (i.e. a so-called full mesh conference), however this is a theoretical solution for which many practical impediments remain.

Large scale UC cloud services have yet to come online, although trials have started from the major vendors (e.g. Cisco and Microsoft).  Analyst forecasts on the adoption of UC in general and hosted UC in particular are high; however forecasting adoption rates is notoriously difficult.  The building of new data centers has a long lead time; scaling up existing data centers is faster but, as discussed above, over-provisioning a data center has unintended consequences that will impact the quality of service.  Efficiently and effectively scaling a cloud UC service will require a careful balance of strategic placement, good forecasting as well as the creation of an optimal overall architecture that addresses all the issues that will be encountered through the growth period of cloud-based UC.

Routing Challenges

The current routing method for SIP, like email, involves the Domain Name System (DNS).  When SIP servers get a call set-up request, they perform a DNS query to resolve the human readable domain name (e.g. google.com) to an IP address (e.g. 74.125.224.18).  To differentiate between, for example, SIP related queries vs. email related queries, DNS provides domain sub-records (known as SRV records) to provide the IP address of the different servers within the domain that handle different types of requests.  This works well for a large enterprise with self-hosted UC, the SIP server is able to resolve any SIP address within its own domain.  A small business that outsources its UC service can employ an SRV record to redirect UC traffic to the cloud.  However, the hybrid scenario (e.g. a large multi-national that self-hosts it’s UC for corporate HQ, but chooses to have its international subsidiaries hosted locally) presents a problem.

A SIP request for a specific employee would need to resolve to a specific data center which may be hosted by different vendors in different parts of the world, e.g. Telefónica, France Telecom, Tata Communications, etc.  However, a generic SIP address (e.g. sip:jose@acme.com") would not give a clue as to the location of the employee and therefore which SIP server would be able to route the call.

One resolution would be sub-domains, e.g. ‘cala.acme.com’, which could be more manageable.   Another solution would be if companies employed the appropriate top level domain country prefix, e.g. sip:jose@acme.br (assuming that domain is available), for someone based in Brazil.  However, there is no guarantee that the correct top-level domain or sub-domain would be provided by the calling party.  Therefore, the only remaining option would be to provide a DNS record per employee.

While large corporations tend to manage their own DNS name servers for other reasons and clearly have a record of each employee’s location, relying on DNS to resolve individual SIP URIs would not work because the DNS schema does not extend to resolving the user portion of a SIP address (i.e. the part before the ‘@’).  An extension of DNS, the ENUM system could accommodate this.  However, this would require an extension of the ENUM standards, as well as the creation of a global directory of individual SIP addresses.  Since firms tend to not publish their corporate directories online, there will be a reluctance to do so in an ENUM-type directory which can be accessed by telemarketers, spam bots and other undesirable entities.

Regulatory Implications

Accommodating IP Communications in Telecom Regulations

Sovereign governments world-wide have long imposed regulation on various forms of communication, both to protect the public and national interests and also as a basis of taxation.  Due to the rapid emergence new forms of communication based on the internet, these regulations have often failed to keep up with innovation.  Where regulation does cover next generation communications, it is often an adaptation of existing regulation that does not take into account the unique attributes of the new technology.   An example of this in the US is the extension of the 911/emergency calling regulations to IP telephony, which has severely hampered the adoption and deployment of this technology.

The adoption of enterprise UC has confirmed the latent demand for these services.  However, since UC is still in the ‘early adopter’ phase of the technology adoption lifecycle, most of the current inter-enterprise communications continue to be made via the PSTN (i.e. UC sessions converted to PSTN calls by an IP gateway).  This will continue until critical mass is achieved and the network effect causes UC-enabled firms to start to conduct a significant proportion of their communications over the internet, thereby bypassing national regulations.  At that point the national regulators will have two options:

  1. Ignore UC, as they have tended to do with email and consumer services such as Skype;

  2. Put in place regulations on UC analogous to the regulations on the PSTN.

It is hard to predict what 190+ national governments will do in the face of rapidly evolving technology.  However, learning from the E911 debacle, it seems clear that unless regulation is put in place before large scale deployment, governments, vendors and customers will be placed in the difficult situation of complying with post-hoc regulation that could negate some of the benefits that UC could have on the economy.

Maintaining National Jurisdiction over Communications

Given that communication networks are a strategic national asset, it seems unlikely that many sovereign governments would choose the laissez-faire option.   Quite apart from anything else, government derives its power from legislation; so failing to legislate to accommodate a major communications revolution would be a delegation of that power.  Unless UC is regulated, there will be several areas of public interest that will be negatively affected.

Tax revenues

Since UC is, as yet, unregulated, it seems clear that tax revenues from communications will erode as firms start to abandon wireline PSTN services in favor of UC.  All firms are subject to corporate taxes, so if the telephony incumbents are superseded by other firms, that will not affect tax revenues.  Indeed, to the extent that UC services add value to communications and drive new revenue, the tax revenues related to communications will increase as UC becomes a mainstream medium.   However, governments also derive significant tax revenues from the consumption of communications, so that revenue will have to be replaced.  One method much simpler than extending various telephony taxes would be making UC service subject to existing sales taxes.

Lawful intercept

The enhanced communications features of UC are particularly attractive to those who operate outside the law, given that IP communications enables encrypted communications.   Also, because UC media packets flow over arbitrary routes, they are difficult to intercept.  It may be that more developed nations have already found ways to intercept and decrypt UC communications over the public internet.  If that is the case, then it may not be in their national interest to create lawful intercept regulations for UC, as it would give lawbreakers something to try to bypass.   However, for less developed nations, it seems clear that they must develop effective ways to monitor UC traffic.  As discussed in the ‘Scaling issues’ section of this paper, if UC media is required to traverse a fixed intercept point (i.e. the data center), UC calls would be subject to additional latency and that would affect the quality of service.

National security

Communications networks are a national strategic asset. Traditional communications technologies are domiciled within the region that they are used and therefore they are subject to regulatory oversight as well as being physically protected within national borders.  However, cloud based UC enables UC service to be delivered from data centers domiciled beyond the national borders, indeed it would be more economically viable to do so for a small country.  However each sovereign government’s perception of its national interest is likely to over-ride economic goals; so cloud UC vendors may be compelled to domicile data centers within national boundaries in order to be allowed to offer service in those countries.

Business Implications

Metcalfe’s Law and Network Effect

Most of the business implications of cloud based UC are affected by Metcalfe’s Law, so we should start with a discussion of that.  Robert Metcalfe formulated this notion to describe the so-called ‘network effect’, i.e. that a communications network becomes more valuable when more people (or devices) start to use it.  This was expressed mathematically as n(n-1)/2, where ‘n’ is the number of users or devices on the network.   However, if n is a very large number (say 10,000,000), this is not to say that each of us has 10m contacts; but it becomes more likely that all of our contacts are included in the 10m.  This is where network theory enhances Metcalfe’s Law.

Clearly, Metcalfe’s Law and network theory have accurately predicted the development of the internet, and the various services that can be found on the internet, including email and instant messaging as well as the social network services such as Facebook and Twitter.  The adoption of UC is also subject to Metcalfe’s Law and network theory, although some UC modalities are affected differently to others.

UC to UC voice communications can take advantage of an entirely different network (i.e. the PSTN) to significantly enhance the UC network effect by leveraging the PSTN network effect.  Gaining access to the PSTN was a key determinant for UC gaining any kind of adoption.  An interesting side note is that this proves a key element of Metcalfe’s Law in that small networks do gain disproportionately by attaching to large networks.

On the other hand, the more esoteric UC modalities such as data collaboration and application sharing (and to some extent video conferencing) are subject to the more negative aspects of network effect because no industry standards exist and they therefore tend to be developed in a proprietary fashion.  Even between vendors that offer similar features, these services are not interoperable and therefore customers’ utility of these features is limited.  This will impact the establishment of these modalities as mainstream forms of communication.

Cooperative Peering among Competitors

For UC to become a mainstream medium, particularly across all modalities, the competing vendors and service providers must establish interoperability.  As previously discussed, UC interoperability is partly limited by a lack of standards.  Where standards exist, they are generally subject to various valid interpretations that limit interoperability.  The creation and adoption of standards and the work required to verify interoperability is basically a function of the competitive strategy of each vendor.

A simple assumption would be that all service providers would gain by establishing interoperability (i.e. a rising tide lifts all boats).  As mentioned above, one of the predictions of Metcalfe’s Law is that small networks gain disproportionately when attaching to large networks.   Herein lies the issue: for every potential bilateral service provider relationship, each party has its own perception of who owns the larger network and therefore who stands to gain most.  It is precisely this issue that has prevented the interoperability between the various instant messaging services; so if this example is any guide, there will never be interoperability between UC clouds.  However, there is hope on the horizon.

One way to break the Metcalfe Law deadlock is via regulation.  It is clearly in the public interest to ensure that all communication networks will interoperate and there exist regulations to ensure that telephony service providers provide inter-network communication.  With the recent global financial crisis we have come to learn that, while markets are efficient in general, some natural market inefficiencies must be proactively managed and UC is a classic example.  There are major gaps in the standards and interoperability related to many modalities of UC, including:

  • Presence and instant messaging;

  • Videoconferencing and ‘Telepresence’;

  • Data collaboration and application sharing.

Whether or not regulators choose to micro-manage all aspects of UC communications, UC service providers must be incented to facilitate interoperability.  The rising tide will lift all boats, but if the boat owners were left to control the tide by general consensus, there would be no tide at all.

Another way that the Metcalfe Law deadlock will be broken is via corporate self-interest.   Once it becomes very clear that, unlike free consumer instant messaging, cloud-based UC has enormous commercial potential then the vendors will be compelled to seek ways to realize that potential.  Email standards rapidly converged after an initial period of competing standards and technologies, and it is likely that this will happen with UC.  Indeed, there are already fairly robust bilateral interoperability programs being run by the major UC vendors and also some multi-lateral programs (e.g. the UCIF).  However, it is true that SIP Trunking interoperability has been slow to emerge and this is partly due to the incumbency of the telephony network service providers.

Risks of Regulation

One aspect of the deregulation of the telephony business that started in the 1980s was that the manufacture of telephony equipment was spun off into businesses that were separate from those of providing telephony service.  This gave rise to companies like Lucent Technologies that were free to innovate outside the competitive strategy of their former parent companies and to compete in international markets.   Interestingly, much of the UC innovation has been done by firms other than the traditional telephony equipment vendors (e.g. Cisco, Microsoft and Skype) and it is these firms that are forging ahead with ambitious plans for UC in the cloud.  This is a natural extension of their enterprise UC business strategy, i.e.:

  • To capture the 80% of the total addressable market that is represented by small and medium sized businesses;

  • To drive UC adoption within the enterprise market segment by UC-enabling the small and medium firms;

  • To ensure that they don’t yield the small and medium segment to other vendors who would then use that base to encroach on the enterprise segment.

So getting into the UC cloud business makes sense for the enterprise UC vendors, but it also has risks.

Earlier in this paper the subject of the regulation of UC service was discussed.  If various sovereign governments declare UC cloud networks to be a substitute for telephony and therefore within the remit of telephony regulation, then there is the risk that the UC vendors themselves will become regulated entities.  Once again, it is not possible to predict how 190+ sovereign governments will react to the advent of cloud-based UC networks.  However, the notion that a technology company would become regulated as tightly as a telephony service provider is clearly not something that would be welcome in the fast moving technology industry.

However, there are mitigations to the risk of regulation.  UC vendors can choose to:

  1. Isolate their main business from regulation by spinning off separate UC cloud service companies;

  2. Elect to enter or not enter various national markets where the regulatory environment has the greatest risks (e.g. Google’s decision to withdraw their search service from China);

  3. Enter into joint ventures or licensing agreements with incumbent telephony service providers (i.e. Cisco’s current strategy);

  4. Lobby the national governments to attempt to manage the regulatory impact.

Whichever strategy or combination of strategies that the UC cloud vendors adopt, there will be a cost or revenue impact other than if they were to remain unregulated.

UC Vendor Coopetition with Telephony Incumbents

As discussed above:

  • UC is a revolutionary technology that seems certain to supersede the traditional communications networks;

  • UC is currently gaining ground partly by leveraging the global connectivity of the PSTN;

  • The current UC equipment vendors are not telephony service providers;

  • There are regulatory risks of entering into the UC cloud service provider business.

Therefore the rise of UC has set the stage for an interesting period of competition between the telephony incumbents and the UC vendors.  While landline PSTN usage is in decline among consumers, the telephony incumbents are, for the most part, focusing on mobile telecommunications revenues to fill the gap.  If, as seems likely, UC starts to erode business usage of the PSTN, then the incumbents will seek ways to replace that business.

One replacement strategy would be for traditional service providers to offer cloud-based UC as a new line of business.  This would be synergistic with:

  • Their current ownership of the relationship with business customers;

    • This is not to say that these same business customers are not also customers of the UC vendors, it is just that the relationships are often with different parts of the customer organization (i.e. telephony procurement vs. network or applications procurement).

  • Their current relationship with national governments who see communications as a strategic national interest;

  • Their ownership of ‘last mile’ of communications, particularly for the provision IP/MPLS networks;

  • Their ownership of large portions of the ‘internet backbone’;

  • The need of UC to gain access to the PSTN and mobile radio networks for the foreseeable future;

  • UC vendors’ desire to avoid regulation.

Maximizing shareholder’s return on equity is often at odds with sharing new business opportunities with, or seeing old business being encroached by, competitors.  Without doubt both the telephony service providers and the UC vendors view their own success as a zero-sum game.

A precursor to the competitive struggle mentioned above is currently being played out in the provision of SIP Trunking service to enterprise UC installations.   Telephony service providers are keen to:

  • Ensure that inter-company UC voice traffic remains on their telephony networks;

  • Have their business customers purchase more bandwidth on their IP network connections.

Where both calling parties have access to UC, the alternative to a SIP Trunk call would be to make the call as a toll-free, wideband connection over the internet as either a UC ‘federation’ session or as an enterprise-to-enterprise SIP Trunk; so SIP Trunking itself is not the best option for the customer.  For the telephony service provider, the provision of ‘SIP Trunks’ simultaneously:

  1. Cannibalizes PRI revenues because the pricing model for IP connectivity is often cheaper on a per call basis than a PRI trunk;

  2. Facilitates the adoption of UC systems by their customers which:

    1. Does not natively use telephony network technology;

    2. Empowers new competitors and highlights the value of substitutes to telephony.

The result of this situation is that telephony service providers are deeply conflicted by SIP Trunking: the sales organizations want to sell SIP Trunks but the network organizations place obstacles in the path of fulfillment (e.g. by declining to participate in the creation of, or adhere to, standards for SIP Trunking).

Despite these conflicts, the forces that are described above, particularly those of incumbency and regulation, are sufficiently powerful to cause alliances to be created between telephony service providers and the UC vendors.  Should the UC cloud vendors decide to compete with the incumbents, the outcome will most likely be decided by the resolution of the ‘net neutrality’ debate.

Net Neutrality

Much has been written about net neutrality of late and it doesn’t bear repeating here.   However, it seems clear that UC cloud traffic will be carried over IP networks that are unlikely to be owned by the UC cloud service providers.  This creates several conflicts of interest:

  • It is likely that the last mile/MPLS network provider will also offer cloud-based UC and will rail at the prospect of carrying UC traffic for a competitor, resulting in:

    • Revenue lost to the competitor;

    • Additional cost of provisioning and managing the network for limited incremental revenue.

  • Given the high bandwidth utilized by UC services, it is likely that quality of service for all customers will be impacted on under-provisioned network links, either by:

    • The UC cloud customer having limits placed on their bandwidth utilization;

    • Other customers seeing network congestion impact their utility of the network.

Clearly the win-win for all parties would be the existence of virtually unlimited bandwidth.   However, the capital cost of achieving that goal world-wide is prohibitive; and is therefore unlikely to be realized in the short to medium term.  It is not in the public interest for:

  • Either UC cloud vendors or IP network providers gain at the other’s expense;

  • The adoption of new and economically beneficial technologies (such as UC) be restricted by IP network turf wars.

Yet without regulatory intervention, it is naïve to assume that these outcomes will not materialize.

Connecting with Consumers

In the same sense that collaborative peering between UC cloud services for small and medium sized businesses will benefit the general adoption of UC, so also would collaborative peering between the UC clouds and the Consumer UC services.  The advantages and the apparent disadvantages of doing so have also been discussed.  However, the main impediment to the consumer networks being open to collaborative peering is the lack of an effective business model for network peering that would offset any currently perceived disadvantage.

Microsoft Corporation now enables multi-media federation between Lync and Windows Live Messenger.  This agreement was clearly easier to facilitate within the same company than between two separate firms whose goals may not be aligned.  If this service is successful, it will serve as a test case for the benefits of collaborative peering between consumer services and commercial UC services.  Ultimately, free (or indirectly funded) consumer service providers will be open to similar peering agreements if:

  1. The lack of federation with other communications services causes a migration of users to other namespaces that offer this feature;

  2. Consumer service providers can find a way to monetize federation with cloud or enterprise UC (e.g. a payment model similar to ‘800’ service today).

Summary

The success of UC technology in the enterprise has clearly presaged the expansion by UC vendors into the small and medium business market segment.  The inclusion of smaller businesses and, ultimately, consumers in the UC user experience will be to the benefit of all.  The specific needs of the SME segment can best be served via a hosted service model and this has created the opportunity for the major UC vendors to consider UC as another service that can be provided from ‘the cloud’.  While cloud-based UC could be the next multi-billion dollar technology business, there remain many issues to be resolved before this goal can be realized:

  • There are significant technology challenges to be overcome, i.e.:

    • Latency mitigation;

    • Scaling to cope with rapid adoption and massive bandwidth consumption;

    • Finding new ways to route SIP messages.

  • Communications regulations and issues of national interest must be addressed in many countries around the world;

  • Business strategies must adapt to market realities and new alliances must be created to ensure that the adoption of UC technology is not impeded.

With the successful deployment and adoption of cloud-based UC, as well as integration with consumer IP communications services, the UC revolution will essentially be complete.


If you liked this article, please comment, share and/or rate it below.   If you didn't like it, please comment!

Comments powered by Disqus+