What Is Cloud Computing?

Cloud computing is a model for enabling on-demand network access to a shared pool of configurable computing resources, including networks, servers, storage, applications, and services, that can be rapidly provisioned and released with minimal management effort or service provider interaction. The definition, formalized by NIST in Special Publication 800-145, identifies five essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service. This last characteristic reflects the utility computing model that underpins cloud economics: consumers pay for resources in proportion to what they use, rather than provisioning capacity for peak demand. Cloud computing draws its foundations from distributed computing, grid computing, and virtualization technology, extending those paradigms with multi-tenant infrastructure, standardized programming interfaces, and large-scale automated resource management.

Service Delivery Models

Cloud computing is organized into three standard service layers. Infrastructure as a Service (IaaS) provides virtualized compute, storage, and networking, giving consumers control over operating systems and deployed software while the provider manages the physical hardware. Platform as a Service (PaaS) adds a managed application runtime, database services, and middleware above the infrastructure, allowing developers to deploy code without configuring servers. Software as a Service (SaaS) delivers complete applications, such as email, office productivity tools, and customer relationship management systems, directly to end users over the web. Web services and service-oriented architectures form the technical substrate connecting these layers: REST and SOAP interfaces expose cloud functionality in a provider-agnostic way, and service computing research examines how to compose, discover, and govern these services reliably at scale.

Infrastructure and Virtualization

The operational efficiency of cloud platforms rests on virtualization, which allows a single physical server to host many isolated virtual machines or containers. Hypervisors partition CPU, memory, and storage among tenants while enforcing isolation guarantees. Network Function Virtualization (NFV) extends this principle to networking, replacing dedicated hardware appliances, such as firewalls and load balancers, with software running on commodity servers. Software Defined Networking (SDN) complements NFV by separating the network control plane from the data plane, allowing network topology and routing policy to be managed programmatically through a central controller. Together, NFV and SDN give cloud providers the flexibility to reconfigure network infrastructure in software on timescales of seconds rather than the hours or days required to provision physical hardware. Research on fronthaul-constrained cloud radio access networks in IEEE Wireless Communications illustrates how these virtualization principles extend into telecommunications, where baseband processing is migrated from physical base stations to centralized cloud pools.

Distributed and Edge Extensions

Cloud computing did not replace earlier distributed paradigms; it subsumed them and extended outward from centralized data centers toward the network edge. Grid computing, which federated geographically dispersed computing resources for scientific workloads in the 1990s and 2000s, addressed many of the same resource sharing problems that cloud platforms solve today at larger scale and with simpler user interfaces. The NIST Cloud Computing Standards Roadmap (SP 500-291) documents the standards landscape for interoperability and portability across cloud providers, identifying gaps that inhibit workload migration between platforms. Edge computing moves processing closer to the data source, reducing the latency and bandwidth costs of sending all data to a central cloud. Dew computing carries this logic further, placing persistent services on local devices so that functionality is preserved even without a network connection. These edge and dew tiers increasingly operate as complements to the central cloud rather than replacements for it, with workloads allocated to whichever tier minimizes cost and latency for a given application.

Applications

Cloud computing has applications in a range of fields, including:

• Big data analytics and machine learning, using elastic compute clusters for large-scale data processing
• Internet of Things platforms, aggregating and processing sensor data from distributed device fleets
• Enterprise software delivery, through SaaS applications replacing on-premises installations
• Scientific research computing, with burst capacity for simulation, genomics, and climate modeling
• Virtual power plants and smart grid management, coordinating distributed energy resources through cloud-hosted control systems