The "cloud" is aptly named. It seems as amorphous as a sky of scattered cumulous on a windy day. DE has collected the glossary of terms below to help give it definition.
"Cloud computing is 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," according to the National Institute of Standards and Technology (NIST).
That's a pretty good textbook definition. NIST goes on to say that cloud computing has five essential characteristics:
- On-demand self-service.
- Broad network access, including access via thin clients.
- Resource pooling, in which storage, processing, memory, network bandwidth, and virtual machines are serving multiple users.
- Rapid elasticity, meaning the ability to quickly scale.
- Measured service, which refers to the services being metered in some way -- often via storage or bandwidth.
The elasticity point is key. Gartner, a research firm, calls it out in its definition as well. To quote the company's Daryl Plummer, managing vice president and Gartner fellow, cloud computing is "a style of computing where scalable and elastic IT capabilities are provided as a service to multiple customers using Internet technologies."
Delivering the Cloud
Bob Williams, Autodesk product marketing manager for simulation, says that you will hear two primary terms mentioned.
Cloud computing refers to actually using hardware on the cloud for numerical crunching, processing, etc.
Software-as-a-service (SaaS), which is more about the delivery method, is the way that you use and gain access to the software. It's a service model.
"Sometimes those two factors are combined together," Williams explains. "You may be using an SaaS version of software that's also doing all the heavy calculations using the cloud, but they don't have to be connected."
Two other service model variations are infrastructure-as-a-service (IaaS) and platform-as-a-service (PaaS), which allow users to upload their own software to use on demand. NIST differentiates the two by limiting IaaS to provisioning "processing, storage, networks, and other fundamental computing resources, where the consumer is able to deploy and run arbitrary software." It defines PaaS as allowing cloud deployment of "created or acquired applications created using programming languages and tools supported by the provider."
Some definitions mention utility computing as a synonym of cloud computing. Utility computing is an older term used to describe on-demand access to storage and virtual servers.
Different Types of Clouds
Looking from another angle, ANSYS developers think of cloud computing as any remote access to a computing infrastructure required for the effective use of simulation software. As such, the term includes a private cloud, such as enterprise customers with centralized high-performance computing (HPC) resources and distributed users; hosted cloud, in which remote dedicated HPC hardware is hosted by a third party, such as SGI Cyclone; and public cloud, such as Amazon EC2 and Microsoft Azure. Hybrids are also possible.
Hopefully that clears up cloud computing.
Contributing Editor Pamela Waterman, DE's simulation ex- pert, is an electrical engineer and freelance technical writer based in Arizona. Jamie Gooch is DE's managing editor. You can send them e-mail to firstname.lastname@example.org.