The analogy between the evolution of the electric energy industry and cloud computing is oftentimes used, and for good reason. It’s likely the most applicable predictor of where this industry is heading over the next 10-20 years.
Although slight regional variances exist, it’s generally the case that I, as a consumer of electric power, can plug in my appliance anywhere in the world and expect it to work efficiently, safely, and reliably. Standards for voltage regulation, plug/outlet design, and circuit protection are mature and widely embraced, and the electric appliance industry can compete, and innovate, on a level playing field for the benefit of consumers worldwide. The clock radio in my office is one of thousands of such devices in the market today. It is a Japanese brand, with technologies patented in the US and subcomponents manufactured in South Korea. This amazing device, purchased for US $12, is incredibly reliable and runs on “pennies” of electrical power per week.
The emerging global cloud computing market space, built on a fabric of underlying standards promoting interoperability, security, assurance, and trust, also promises to amaze consumers in the coming decade. Current computing standards aside, the paradigm shift to the cloud compels us to address new areas of standardization where none have existed before. These include the following:
- Industry-wide agreement and widespread adoption of standard terminology and measures for basic elemental units such as “compute instance,” “compute-hour,” “MB-month,” and so on
- Standard language and provisions guiding the overall structure and framework for a new generation of consumer-friendly cloud computing service-level agreements (SLAs)
- Foundational standards for portability and interoperability of compute instances (virtual machines) across and among participating public cloud computing service providers, as well as private cloud architectures serving an enterprise
Cloud Computing 2020
One might imagine a not-too-distant future where consumers of standards-based cloud computing services are able to automatically and rapidly (i.e., in near real time) provision and re-provision their enterprise computing load across multiple competing providers based on a number of attributes, including instantaneous pricing, geographical region, temporal (time) shifting, service-level requirements, and so on. Standards will allow consumers to seamlessly move their computing load from their private cloud (enterprise data center) to public cloud providers and back again, as demand or failover requirements dictate.
In the future, we might even see a virtual “stock market” develop for brokers or aggregators of cloud computing services — where compute instances, capacity, and options are traded, in real time, much like stocks are traded on a stock exchange. Enterprises that continue to invest in “bare metal” data center capacity and subscribe to industry cloud computing standards for interoperability may even have the opportunity to sell excess capacity back to the “grid,” in much the same way that a homeowner with a residential wind power generator can sell back excess power to the grid in utility markets that provide “net metering” capabilities.