Cloud computing is one of the most talked about trends in information technology today and it has already significantly impacted the IT landscape. Major players such as Microsoft, Google, and Amazon are making massive investments in cloud computing infrastructure as they make serious financial bets that cloud computing will continue to gain market share. One can argue cloud computing first began during the days of timesharing on mainframe computers in the 1960s and ’70s. Cloud computing has evolved into what it is today through the influence of various concepts and technologies, such as grid computing, utility computing, software as a service (SaaS),
application service provider (ASP), virtualization, multi-tenancy, etc. Ultimately, cloud computing is a culmination of experiences and advances in areas such as clustering, connectivity, abstraction, convention, and culture. The following section reviews some of these factors in more detail.
Clustering
Key to the cloud model is the reliable provision of computing cycles. How much computational power can be packed into a square meter with minimized cost? A few decades ago, integrated circuits revolutionized the production and performance of electronic equipment. The revolution from discrete electronic components to integrated, cheap chips led to enormous improvements in the efficiency of computing operations and allowed logic to be integrated in the range of devices we see today. This same densification has continued in the computing world with multiprocessor and multicore processors and continues further with clustered racks of blade servers in data centers and now with integrated containers of computing equipment or modular prefabricated data centers to further enhance performance, efficiency, and automation. This clustering effect provides compute cycles with a critical mass and a cost per compute cycle point that traditional IT deployments cannot match.
Connectivity
Where is the most efficient place to perform the computing task? Connectivity enables effective deployment of computational power on all levels in an application architecture. Each discrete element has to be connected in some way with the rest of the system. Fast, reliable, cheap interconnectivity is the cornerstone of various communications types – whether on chip, a single computer unit, within or outside the data center. The Inte
et has revolutionized the way we utilize information, access application functionality, and perform social interactions. Ubiquitous network connections enable access to cloud computing from a variety of applications, devices, and workflows. Mature connectivity allows us to delegate tasks to specialized applications in the cloud that can use the full power and elasticity of the cloud computing infrastructure.
Abstraction
Modern cloud architectures simplify access to computing resources. To what degree do we need to deal with IT infrastructure details to solve business problems or execute operational tasks? Abstraction enables flexibility and facilitates the right scale of resources to operate on a given problem. In the software industry, abstraction has been achieved by various technologies and architectural designs, most notably and recently service oriented architecture (SOA) and virtualization. SOA is an architectural style to abstract business functions away from the concrete software provider and a style for scalable, distributed computing, while virtualization enables users to fully decouple runtime applications or services from the dependency on the underlying hardware infrastructure. Cloud computing typically uses virtualization inte
ally for scaling, elasticity, and reliability, while applications delivered to the end users use SOA.
Convention
What is the right ratio of uniformity versus specialization? Conventions or standardization helps to decrease unnecessary variability and improve efficiency in manufacturing and integration. This leads to cheaper components and to commoditization of the base IT elements. Cloud computing leverages commodity hardware and standards in communications and application delivery. This enables economical operations to run in the cloud on the provider side, as well as easy access from the user or software vendor side. A typical cloud environment runs on cheap but highly redundant hardware infrastructure that enables high reliability. More specialization is occurring on the software side.
Culture
Are we comfortable with our processing and data existing in the cloud? During the past decade we have been conditioned to take ubiquitous access to the network for granted. Moreover, Inte
et, social media, Web applications, Web mail, Web storage, peer-to-peer networks, videoconferencing, and many other styles of communication have become popular and widely accepted by consumers. Microsoft, Google, Amazon, and others use cloud computing to deliver these experiences. Organizations are now beginning to take advantage of these cloud computing concepts to deliver on their mandates.
Evolution in Data Center
Traditional data centers, similar to electronic components in the early stages of computing, typically feature individual servers that are isolated, underutilized, and expensive to maintain. This leads to high power consumption, physical space constraints, and a high cost to provision the full data center, especially for mission-critical systems that require redundancy and failover protection.
Cloud Computing Definition
Cloud computing is a next-generation software (application and services) hosting technology that can be owned and operated by an organization (the private or inte
al cloud) or by independent provider (public cloud).
Cloud computing uses the latest innovations and experiences across computing technologies to maximize efficiency of operations, while minimizing the cost of computing components, electrical power, space, and time required to provision additional resources. Efficiency is reflected in typically using commodity components and uniformity with imposed conventions and standards that might be delivered as a “cloud computing appliance” that ties together pre-integrated and pre configured self-healing computing systems. This dramatically simplifies maintenance, reduces the need for on-site visits, and provides turn-key solutions for provisioning additional cloud infrastructure resources. All interaction with the cloud infrastructure is done through Web services or self-service Web portals for unified management. Clouds can host heterogeneous applications, provide multi-tenancy, and precisely measure resource usage per application.Cloud computing uses highly parallelized virtualized computing infrastructure with fully automated operations to provide failure resiliency, load balancing, and agile elasticity in scaling up and down without assistance from IT personnel. This makes it possible to dynamically and on demand allocate or deallocate new computing resources for running business applications.