Clearing the Air on Cloud Computing
A new discussion document by Will Forrest from McKinsey & Company focuses on setting realistic expectations on Cloud Computing by taking a "hype free" approach starting with the most basic question of what a "cloud" actually is.
While it has great potential, many of the claims being made about cloud computing have lead some to the point of "irrational exuberance" and unrealistic expectations.
The report starts with the attempt to propose an industry-standard definition of cloud computing (similar to my previous post) including what makes it unique. According to Will, there are currently over 20 definitions of Cloud Computing, most of which do not distinguish cloud services from clouds themselves, do not explicitly emphasize abstraction of infrastructure and do not provide definitive economic implications of Cloud Computing. In an attempt to overcome the shortcomings of the current definitions, Will proposes the following one:
Clouds are hardware-based services offering computing, network and storage capacity where:
- Hardware management is highly abstracted from the buyer
- Buyers incur infrastructure costs as variable OPEX
- Infrastructure capacity is highly elastic (up or down)
He also defines the following Cloud Computing characteristics, closely related to his definition:
- Enterprises incur no infrastructure capital costs, just operational costs and operational costs are incurred on a pay-per-use basis, with no contractual obligations.
- Architecture specifics are abstracted. In addition, run in multi-tenancy mode with multiple users accessing the infrastructure simultaneously
- Capacity can be scaled up or down dynamically, and immediately, which differentiates from traditional hosting service providers.
- The underlying hardware can be anywhere geographically.
Such strict definition allows to easily differentiate between a true Cloud (for example Amazon, Google or Windows Azure) and a Cloud service (for example ZOHO, SaleForce.com and Gmail). While a Cloud has to comply with all three key requirements, including the abstraction of the underlying hardware, elastic scaling and per-use billing, a Cloud service only complies with two key requirements - the abstraction of the underlying infrastructure and elastic scaling.
The rest of the report tries to identify the types of customers that should be early adopters - those that can benefit from existing and planned commercial offerings. According to Will:
Clouds already make sense for many small and medium-size businesses, but technical, operational and financial hurdles will need to be overcome before clouds will be used extensively by large public and private enterprises
On another hand Will notes that there are significant hurdles to the adoption of cloud services by large enterprises:
- Financial - Current cloud computing offerings are not cost effective compared to large enterprise data centers
- Technical - Security and reliability concerns will have to be mitigated and applications re-architected
- Organizational - The IT supply and demand organizations will have to adapt to function in a cloud-centric world
- Operational - Business perceptions of increased IT flexibility and effectiveness will have to be properly managed
The report has caused quite a few reactions. Andy Greenberg from Forbes notes:
"The cloud" has come to represent the bright future of computing, a world where processing and storage become as ubiquitous, cheap and accessible as electricity. But for big business, one researcher argues that "cloud" metaphor may be economically apt: The closer you look at the much-hyped technology's price advantages, the fuzzier they seem. Much of cloud computing's misplaced hype, contends Forrest, comes from the assumption that businesses that make the switch will be able to do away with their entire IT department... But in his analysis of McKinsey's financial services client, Forrest found that ... moving to Amazon's service would only cut about 200 full-time workers... of the company's 1,700 or so IT employees... hardly the savings chief information officers might imagine.
Steve Lohr from New Work Times:
... while cloud computing is a marketing triumph, new research from McKinsey & Company asserts that trying to adopt the cloud model would be a money-losing mistake for most large corporations... We should focus on things we know work now, and virtualization works
Finally, Nicholas Carr notes that:
McKinsey, of course, has an interest in encouraging companies to maintain large and complex in-house IT operations... Nevertheless, the McKinsey analysis is a valuable one, not least because it underscores how early we are in the development of the utility-computing grid - and why we shouldn't expect large companies to begin shutting down their data centers any time soon... The scenario McKinsey analyzes - the wholesale replacement of a large enterprise computing operation with rented space within an external cloud - is a bot of a straw man. The real opportunity that the cloud offers large companies today is as a supplement or complement to their in-house operations rather than as a complete replacement. The cloud model offers a way to gain access to additional computing and storage capacity, particularly to cover fluctuations in demand or carry out a short-term data-crunching exercise, without having to make capital investments in new equipment or hire more workers. The cloud also, of course, provides a way to tap into powerful software-as-a-service applications that can provide substantial savings, not only in equipment and labor but in licensing and maintenance fees, over the cost of installing an in-house application.
The report itself and the reactions to it are a proof that we still do not have enough knowledge/data about correct usage and economy of Cloud Computing.