Oligopoly: A market in which there are a limited number of providers providing the same service. Its political counterpart, oligarchy, means rule by a few.
Is the cloud computing marketplace becoming the domain of a few big vendors? With large players including Amazon, Microsoft, Google and IBM coming online with similar types of services, we may be starting to see a consolidation of the primary cloud computing market into the hands of a few powerful vendors.
This may herald the emergence of an IT oligopoly, Owen Rogers, senior analyst at 451 Research, suggests in a recent research note. Actually, he goes further to say what is emerging is both an oligopoly and monopoly at the same time. With identical services comes commoditization, and only big vendors that can deliver huge economies of scale with margins will survive in this space.
He adds that perhaps the “oligopoly would be used for bursting of virtual machine requirements, whereas the monopolistic market would be used for performance-based workloads that demand a service-level agreement (such as a database).”
Is this where the cloud marketplace is going?
It’s not surprising that Infrastructure as a Service (IaaS) offerings — raw compute power and storage — would be limited to a few who can effectively compete in what is a high-volume commodity business. To compete in this space, providers need to rent entire counties to build gigantic data centers that can service hundreds of thousands of users 24 x 7.
At first blush, it would seem the barriers to entry to compete as a Software as a Service (application) vendor would be relatively low. But the cloud requires that software developers now not only be good software developers, but also data center operators as well. And these developers aren’t going to attempt to build and run their own data centers — they’re going to lease capacity from the big IaaS vendors. So again, the path leads to the big, well-financed providers who can compete in a commodity marketplace.
Rogers predicts that the pricing for cloud services from the oligopoly group will need to reflect supply and demand — that is, if a data center is running close to full capacity, user prices will go up. Likewise, a data center most idle will mean low unit pricing for cloud consumers. There’s actually quite a bit of risk for cloud providers, as they need to be able to forecast demand and invest appropriately to meet it.
Rogers doesn’t factor in corporate data centers that are underpinning private clouds, which could also potentially be extended out in a more public way to partners and customers. Statistics often show that corporate servers are only utilized at a fraction of capacity, and therefore leasing out excess capacity as compute services (even if its to another unit to the same business) — or as less-commoditized SaaS — would merely be icing on the cake for many of these organizations. Perhaps a grid-type marketplace could develop to handle the buying and selling of such capacity.