In our previous cloud computing tutorials briefing Cutting through the Hype with Cloud Computing Tutorials: Part 1 Defining Cloud Computing presented a now commonly accepted definition of cloud computing developed recently by the National Institute of Standards and Technology (NIST).
NIST Defines cloud computing as “ a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model promotes availability and is composed of five essential characteristics, three service models, and four deployment models.“
Exploring the five essential characteristics of the cloud computing model enhances our understanding of cloud computing. It provides criteria for assessing vendor offerings and proposed or existing cloud computing implementations.
Here in Part 2 we’ll explore those five characteristics NIST Cloud Computing Standards Roadmap – Version 1.0 considers essential in the cloud model. On-demand self-service, broad network access, resource pooling, rapid elasticity and measured service comprise the five essential characteristics of the cloud computing model.
On-demand self-service may sound more like an advertisement for a new fast food restaurant . Perhaps the analogy is appropriate, because it represents an almost natural evolution for ordering and accessing products and services in the 21st century.
If we can get music, books, software virtually at the push of a button, then why not server time and network storage. In more technical terms, the first characteristic of the cloud computing model, automatic provisioning of computing services without the need for human interaction with those that provide the service.
Broad network access extends beyond just the communications network. It also includes the devices used for connecting to resources on the network. The network facilitating broad access in today’s world – the internet or a private intranet.
Access that network using standards that support diverse types of devices, including smart phones, tablets, personal computers and others. That makes it accessible from any location 24/7 from just about any device. This represents the level of pervasive, ubiquitous access envisioned by the second characteristic of the cloud computing model – broad network access
The third characteristic of the cloud computing model resource pooling requires the sharing of storage, memory, bandwidth, processing and other resources. Similar to the management of a large apartment building, a service provider uses a multi-tenant model to dynamically allot virtual and physical resources dynamically to multiple customers, based on their need.
In the cloud computing model the tenants don’t know their street address let alone their apartment number. The consumer typically releases control and even knowledge regarding the precise location of resources – hence the term cloud. In practice the consumer in some cases may select the country, geographic region or general location of the datacenter(s) involved.
Rapid elasticity the fourth characteristic of cloud systems refers to the speed at which the allocation or reallocation of resources occurs. This gives a consumer the appearance of almost instantaneous limitless resource availability.
The fifth and final characteristic of cloud systems – measured service. It allows the service provider and consumer facilities to transparently monitor, report and control resource use. Basis of metering depends on the service – active users, bandwidth, processing, storage.
This translates into the ability for the consumer to pay based on actual usage of the metered service. Can not resist the taxicab analogy here, because measured service leads to flexible pricing. Taking a long ride to the airport, then pay a set price. Shorter ride downtown pay what is on the meter.
Characteristics one, two and five – on-demand self-service, broad network access and measured service appear pretty straightforward to most people. However characteristics three and four – resource pooling with continuous dynamic allocation and rapid elasticity require more sophisticated enabling technologies.
Virtualization, in existence for decades but now applied to the nth degree provides the foundation for implementing resource pooling with dynamic resource allocation and rapid elasticity.
Before we examining the three service and four deployment models of cloud computing, we’ll explore virtualization in our next cloud computing tutorials briefing “Cutting through the Hype with Cloud Computing Tutorials: Part 3 Virtualization.” So please subscribe for notification when published.
Thoughts, questions or suggestions – comments welcome.