Q2Cloud Computing

Cloud computing systems are defined by a set of essential characteristics, as commonly summarized by the NIST (National Institute of Standards and Technology) cloud computing definition, along with a set of standard architectural components.

Essential Characteristics

• On-demand self-service: a consumer can provision computing resources (server time, storage) automatically, as needed, without requiring human interaction with the service provider.
• Broad network access: cloud capabilities are available over the network and accessed through standard mechanisms, usable from a wide range of client platforms (laptops, mobile devices, workstations).
• Resource pooling (multi-tenancy): the provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with physical and virtual resources dynamically assigned and reassigned according to consumer demand, with the consumer generally having no direct control or knowledge of the exact physical location of the provisioned resources.
• Rapid elasticity: capabilities can be elastically provisioned and released, in some cases automatically, to scale rapidly outward or inward commensurate with demand, appearing to the consumer as essentially unlimited and available in any quantity at any time.
• Measured service: cloud systems automatically control and optimize resource use by leveraging a metering capability appropriate to the type of service, with resource usage monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service, forming the basis for pay-per-use billing.

Major Components of a Cloud Computing System

Front-end: the client-side component through which users interact with the cloud, comprising the user's device and the interface (web browser, dedicated client application) used to access cloud services.

Back-end: the provider-side infrastructure, comprising the physical servers, storage systems, and networking equipment, along with the virtualization layer (hypervisors managing virtual machines), management and orchestration software, and security systems that together deliver the actual cloud service.

Network: the internet or dedicated network connection linking the front-end (client) to the back-end (cloud provider's data center infrastructure), through which all service requests, data transfers, and responses flow.

Service models: the specific way in which cloud capability is packaged and delivered to the consumer — IaaS (Infrastructure as a Service, raw virtualized computing/storage/networking resources), PaaS (Platform as a Service, a managed application development/deployment platform), and SaaS (Software as a Service, complete, ready-to-use applications) — each providing progressively more managed, higher-level functionality with correspondingly less direct control for the consumer.

Deployment models: the arrangement in which cloud infrastructure is provisioned — public cloud (infrastructure shared among the general public, operated by a third-party provider), private cloud (infrastructure dedicated to a single organization, whether managed internally or by a third party), hybrid cloud (a composition of two or more distinct cloud infrastructures, public and private, that remain unique entities but are bound together by standardized technology enabling data and application portability), and community cloud (infrastructure shared among several organizations with common concerns, such as compliance or security requirements).

Together, these characteristics, components, service models and deployment models define the overall functional and architectural framework within which any specific cloud computing system is designed, delivered, and consumed.

Why all five characteristics must be present together: NIST's definition treats these five characteristics as jointly necessary, rather than independently sufficient, criteria for a system to be genuinely considered 'cloud computing' — a system offering only some of these characteristics (for example, a traditional hosted server offering broad network access and measured billing, but requiring manual provisioning by the provider's staff rather than automated on-demand self-service, and lacking rapid elastic scaling) would more accurately be described as traditional managed hosting rather than true cloud computing. This distinction matters practically because it is precisely the combination of on-demand self-service, rapid elasticity and resource pooling together that allows cloud computing to deliver its most valued business benefit — the ability for a consumer to rapidly scale computing capacity up or down to match actual, sometimes highly variable, real-time demand, paying only for resources actually consumed, without needing to negotiate with the provider or wait for manual provisioning each time capacity needs change.

Interaction between components and deployment models: the same front-end/back-end/network component structure applies across all deployment models (public, private, hybrid, community), but the back-end infrastructure ownership and physical location differs fundamentally between them — in a public cloud, the back-end is owned and operated entirely by the third-party provider, shared across many different customer organizations; in a private cloud, the back-end infrastructure (whether physically located on the customer's own premises or hosted by a third party on the customer's behalf) is dedicated exclusively to a single organization; and a hybrid cloud's back-end is split across both categories, with the network component playing an especially critical role in a hybrid deployment, since it must reliably and securely interconnect the organization's private back-end infrastructure with the chosen public cloud provider's back-end infrastructure, in a way that allows workloads and data to move between the two environments as required by the specific hybrid architecture being implemented.