Cloud Computing – Past, Present and Future

June 23rd, 2009 by Julia Lim

Distinguished Analyst and Research VP Tom Bittman breaks down what’s real and what’s not real when it comes to the cloud computing phenomenon in this session, “What You Need to Know about Cloud Computing”.

Cloud computing myths:
1) Only megaproviders will win – False. There are fragmented markets out there with good enough scale for smaller providers and as usual when it comes to technology, innovation makes agility (of a small provider) a competitive advantage. There will be thousands/tens of thousands of providers.

2) There will be a “big switch“ – No despite what Nick Carr wrote, it will be a slow migration that takes decades and won’t be so black and white – cloud and in-house – but instead will end up being more hybrid models (cloud and in-house).

3) Cloud computing is IT commoditization – services offered in cloud may be commoditizing but new applications will be enabled by the cloud.

Evolution of Cloud: Gartner’s been talking about Real-Time Infrastructure for 8 years. Vendors like HP and IBM have talked about “adaptive enterprise“, “on-demand“, “dynamic infrastructure” and more for years as well.

Virtualization is a modernization catalyst and unlocks cloud computing.

• enables economies of scale (sharing) → cloud-ready efficiency
• decouples it from users (culture change) → cloud-ready interface
• speed and elasticity (rapid reaction to change) → cloud-ready speed
• breaks software pricing and licensing (fractional use, consolidated, dynamic change) → cloud-ready pricing
• enables and motivates chargeback (from fixed to variable use) → cloud-ready costing

Once companies virtualize – the #1 benefit is not cost savings but speed.

Efficiency Technologies that enable cloud computing:

• multitenant software
• virtualization
• automation
• parallel computing

Cloud Computing definition again:

A style of computing where scalable and elastic IT-enabled capabilities are delivered as a service to external customers using Internet technologies

Assumption – through 2012 more than 75% of enterprise use of cloud computing will be devoted to very large data queries, short-term massively parallel workloads, or IT use by startups with little to no IT infrastructure. Not a lot of mission-critical. Mainly experimenting with the cloud for now.

How the cloud computing market will evolve:
Today – cloud services tend to be standardized, elastic but in “chunks,” and monolithic in nature. In future, custom service levels, rapid and granular elasticity and open, federated markets of services.

Not everything will move to the cloud. What types of services won’t:

• Not a business differentiator
• Relatively static service
• Very separate
• Service generally not end-customer facing

GOALS:

• self-service, easy-to-use
• independent from business
• ultimate destination is cloud computing

Cloud computing is a continuum: private cloud on one end, public cloud on the other (degrees). Where the cloud services fall depends on ownership and service control/access.  Examples:

private cloud - internal dev/test device
public cloud - web search
in between – business partner cloud services, dedicated SaaS instances (e.g., a dedicated MS Exchange online – “hosted”)

Why Private Cloud:

• low barrier to entry
• elastic and scalable
• lower cost and pay per use
• ease of sourcing migration (stepping stone to public cloud)

Moving to Private Cloud Computing – How to get there:
- service inventory
- service levels/requirements
- current costs for each service
- roadmap for each service
- evaluate and predict cloud service
- business case for private cloud service
- Build: service abstraction and interface, usage metering, shared technology implementation

Popularity: 1% [?]

4 comments June 23rd, 2009