We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.
Start a Discussion
Cloud Computing

Is Cloud Computing in Danger of Becoming Monopolized?

Vote 0 Votes
Is Cloud Computing, as David Linthicum contends in this blog, The danger of the coming 'big cloud' monopolies, heading towards being dominated by just a few big companies? Is this a bad thing, and if so, how can it be prevented?

11 Replies

| Add a Reply
  • I think this is a real possibility, but it depends on what organisations want and expect from the cloud. I beleive there will be monopolies on particular types of applications on the cloud, for example Office. However, I also feel that many organisations like the concepts of cloud computing but wont want to get tied into a big player such as Microsoft or Google. Because of this I see a bigger future (from an organisations point of view) towards private clouds...

    Sure for the general public there will be almost no choice at all, most opting for Microsoft or Google...

  • This is interesting, as my own company is a little guilty of moving to one of the potential candidates for the cloud monopoly. Why? The credibility that comes with a recognizable name is essential for Consected. Its okay for us to be a niche SaaS provider, but if we then rely on a niche infrastructure provider the perceived risk for customers is too great.

    I can't comment on what will make or break the cloud monopoly, but I can make sure that I protect my company from the abuses of any particular provider. Implementing software architectures that leverage cloud resources, without tying themselves too tightly to any proprietary cloud technology, seems like the way to go. As new customers come on board, a SaaS company can start to extend its infrastructure elsewhere, or back in-house if needed.

    Tying the architecture to a specific vendor's cloud computing model is like the past practice of building software for a monolithic mainframe. You can buy additional resources on demand, but you can only go to one place to get them. The cloud is a distributed model, so we are going to leverage that to its fullest extent, from location through provider, and underlying technology.

    Phil Ayres

  • Technology platforms tend towards commoditisation and are therefore ultimately economy of scale businesses. As a result it is inevitable that they will tend towards consolidation and mega-providers. This is just an economic reality as the capital investment required is so large that only a small number of companies can maintain it.

    The issue is for legislators to be proactive about putting a regulatory framework in place to understand the dangers and monitor the cloud marketplace in this context rather than be forced to catch up after it has gotten out of hand. In order to do this they need to a) ensure that there are enough providers by setting an ultimate limit on consolidation to protect competition and b) force providers to give up vertical integration strategies that create walled gardens and limit consumer choice and market innovation. Essentially there are different 'kinds' of businesses that fragment along the lines of infrastructural business models, relationship business models and IP creation business models. In my view any regulatory framework needs to force separation along these lines to enable a competitive market. Taking the telecoms market in the UK as an example the regulator simplistically forced a restructuring of the industry over an extended period so that BT (the provider of the core network) also had to sell capacity to other companies for resale at prices competitive with those they offer their own retail arm. In essence they were forced to separate the infrastructural business from the relationship one in order to stop one organisation having too much power and to foster competition (interestingly something I think needs to be considered for Apple fairly soon - this blog post about whether the ipad impacts 'generativity' also has a discussion of their business model). As a by-product, forcing infrastructural providers to make their platform available to everyone also happens to be in their own interests - the best way to optimise an infrastructural business is to bring it more scale.

    Simplistically then this is what we need to do with cloud platforms - we need to create a regulatory environment in which companies who offer computing platforms are required to do so fairly, with equal access and with no preference to their own branded companies in other business types. The worse case scenario would be for consumers (i.e. companies in this sense) to be locked into one of a small number of walled gardens rather than able to access all of the innovation that springs from an open and competitive marketplace. We probably need to have these frameworks in mind now, however, rather than waiting for the hearings in 5 years that David posits as a possibility.

  • Monopoly no, Oligopoly...probably yes. As most of you probably remember from high school, an oligopoly is where a small cadre of players dominate a market.

    I think that is the case we have today with Amazon, Microsoft, Rackspace, Google and Salesforce dominating the conversation. And given the extreme costs of building out the infrastructure to support the offering, there are few other companies with the resources to play on the scale of the aforementioned companies. But there are, and will be many other players in the market, such as NaviSite and OpSource.

    And that's just the US. There are certainly going to be local and regional players offering cloud infrastructure. This is in part due to local communications and hosting providers wanting to get into the game, but also because of the regulatory and compliance issues around where the data has to be. As an example, Tata Communications recently announced its Instacompute Cloud service starting in India but with plans to extend the services to other markets including the U.S., Europe, Singapore and South Africa.

    And of course there will be private clouds, but those aren't really clouds, are they ;)

  • The way the internet works, attention tends to get concentrated with big names. Unlike brick and mortar businesses, targeting local markets is not much of an option, since the internet is a single huge market. However, under the cloud model, the big names are going to serve the whole market with a mass cloud product, which may be more tuned to the needs of lucrative customers (enterprises). So there are going to be lots of niches which specific needs in terms of product features, or services, that smaller players can target. That is likely to be the model of the market, a few big players, lots of smaller players.


  • The same could apply to the Internet. While some aspects might be closer to be monopolized than others, even the Internet infrastructure is too extended and redundant to become a monopole’s backyard. Cloud Computing is even more so, and there are enough stakeholders to prevent any single one to approach a monopolistic position.

  • At infrastructure level, yes. Oligopoly, actually.

    In fact, the hype and noise around enterprise business has always been dominated by few big players on infrastructure and platform level (and if we include early Enterprise App entrants like SAP, at larger scale apps too). We may not see that part change even with cloud (public, private or hybrid) because of the amount of investments and the larger scale required for larger enterprise needs.

    But otherwise, cloud is a great leveler. We will see lot of niche and smaller scale players compete in specific business problem areas. Combine SaaS and Paas (not iaas type of paas; paas closer to apps) with marketplaces and it really becomes a plausible scenario. It is another matter that we will not hear a lot of people talk about them because of the sheer nature of the niche business. But overall volume covered by these companies will definitely be sizable when put together.

    It is another matter that the hype and noise will always be around big players because that's just the nature of the noise. That's been the case earlier, and will be in cloud, outside cloud, in any industry, in any domain :)

  • We will eventually move towards an oligopoly with smaller players providing specialized niche solutions. If we really are moving to delivering IT as a utility service like electricity, there aren't that many players who will have the muscle (financial and brand) behind them to pull it off.

    We haven't reached the oligopoly state yet - we are currently in the land grab / early mover phase, and then we'll see lots of consolidation and crashes of companies that didn't quite get it right.

    What I question is if the telcos will emerge as one of the oligopoly or be displaced by others. They own customer relationships already - easier for them to add a service as a bundle or upsell an existing customer, than for an unknown to gain a new customer all together. But they seem to be moving pretty slowly...

  • Frankly, David Linthicum's post about 'dangerous' cloud monopolies is a bit alarmist. And I agree with others on this forum that what David really meant was an oligopoly

    Oligopolies are likely to emerge in some Cloud segments (eventually) but not all.

    Any industry that benefits from scale is likely to consolidate over time. This is hardly new. Our on-premise enterprise software industry has gone through much consolidation in the last ten years. So we are likely to eventually see oligopolies in the infrastructure-as-a-service market (currently Amazon, Rackspace etc) which operate on thin margins and need scale to be profitable.

    Another reason consolidation happens is when the value of the product or service greatly depends on number of users and 3rd party support. We saw this with .Net and Java for application development. For similar reasons, we can expect to eventually see oligopolies in the platform-as-a-service market as well (currently force.com, Azure etc)

    We are least likely to see oligopolies in the software-as-a-service segment, because there is much room for specialization and differentiation.

    Even in markets that do eventually consolidate, David's timeline of only five years seems exaggerated. Right now, we don't even have clear business and profitability models. So there is much room for experimentation before things settle down.

  • There will be naturally a slight period of "monopoly" for a few larger providers as the market sorts itself out and comes to a de facto accepted "standard" in terms of management and APIs and capabilities.

    Once that period is over and the de facto standards can be turned into something more industry wide then we'll see growth of other providers to counter the monopolies and drive costs further downward.

  • Doesn't seem far-fetched although 'oligopoly' would be more like it.

    Question is, by merit or (hopefully) not (bec of bandwagon), would it be good for the industry in the long run. Or will there be too much 'trust' for the select few to carefully handle?

    Wishful thinking is the possibility of the 'long tail' if it applies.

Add a Reply

Recently Commented On

Monthly Archives