NIST publishes cloud computing standards to evaluate cloud services

There is no doubt that the most popular marketing term last year was cloud computing, a valuable selling point in both software and service marketing. Some people see this as an "evolution" in the IT field, another technology that is about to be applied on a large scale after virtualization is widely used, and an inevitable outcome of the era of high-speed broadband networks. Others see cloud computing as a managed-based calculation, which is actually a replica of the old model of an input/output device connected to a shared mainframe 50 years ago. Only this time, we have a wider range of applications, even devices such as smartphones and tablets that can connect and share cloud resources over wireless networks for faster performance.

Whether that is the point of view (progress or regression), the change in the entire market brought about by cloud technology is undeniable. Either accept cloud computing or be eliminated, and in various it circles, this view has become a consensus. Refusing to cloud computing's IT career is also seen by some as a short-lived occupation, just as it was in the middle of the 19th century, as a centralized power station emerged, with individual small power supply operators gradually fading out of the market.

If you're planning your IT career, or even further, you're going to build your own cloud computing platform, or compare products from various cloud providers on the market, you need to study rationally and make the right decisions. None of us is willing to make a decision to invest in a cloud-computing project in fear, uncertainty, or a market rambling boast. The fundamental thing that can help us make decisions is a set of cloud computing standards that we can use to measure whether a cloud project or job is right for us.

At present, an independent research authority has given a draft of cloud computing standards, which is the American Institute of National Standards and Technology (NIST). By contrasting the cloud model given by the neutral agency NIST with the product offerings offered by vendors, we can: 1. Determine if the solution meets the minimum standards and workflows for cloud computing solutions, 2. Determine the degree of cloud of the scenario.

To push through cloud computing fog

The NIST cloud definition article can be reduced to one sentence:

Private, public or mixed cloud-with on-demand self-service, broadband network access, responsive resiliency, resource pooling, and measurable services-provides the deployment of architecture, platform, and software services.

If you use this simple definition to compare any existing system, you can confidently determine the cloud value of the system. Solutions that have all of these core elements and are implemented through some kind of cloud and service delivery model have a high probability of success in the marketplace or in your business.

In the NIST cloud definition, as shown in Figure A, a variety of cloud deployment pattern definitions are given, such as public cloud, private cloud. This is the simplest classification, logically based on who owns and operates cloud components such as data centers. It should be noted that the private cloud may be within or outside the enterprise, possibly by an enterprise IT administrator, or by an outsourced service provider, and it is called a private cloud because the architecture serves the enterprise only.

Figure A

In addition to deployment or service patterns, cloud computing contains five core features

Core features of cloud computing service model

Putting aside the deployment patterns used in cloud computing, a cloud solution should deliver services in one of three service modes, the architecture, platform, and services (corresponding to Iaas,paas,saas). These patterns make it clear that the various components of the cloud service provider and the user are divided in the attribution of responsibility. The most user involvement is IaaS mode, with the least participation being the SaaS model.

· In SaaS mode, users consume only software, as if they were running a web email client. Users who have used Google's Gmail or Microsoft's Hotmail mail Service should be able to easily understand SaaS.

· In IaaS mode, users need to assemble and maintain components hosted in the cloud, such as virtual machines, storage pools, firewalls, and other components, which are sometimes called cloud structures (cloud fabric). Amazon Web Service (AWS) and Rackspace are the best vendors for the current model.

· The middle model, PaaS, allows users to deploy applications on the platform provided by the cloud service provider without the need for an administrative architecture. Microsoft's Windows Azure is a typical example of this PAAs model for companies seeking global access and a high reliability platform to build their own application platform.

Once you understand the cloud deployment patterns and service patterns that a cloud solution uses, the next step is to see if the solution covers some of the core features given by NIST.

· On-demand Self Service: According to user needs or automatically to provide users with the appropriate performance, do not require the service provider manual intervention.

· Broad Receptacle Access: Provides localized standard network and Internet access mechanisms for a variety of platforms, such as smartphones.

· Resource pooling:t in some modes, the computers, networks and storage resources hosted by service providers can achieve multiple leases, dynamic resource allocation and resource reallocation according to user requirements.

· Rapid elasticity: The rapid expansion of resources and shrinking capacity, from the user point of view, the service provider has unlimited resources, the enterprise is based on the resources used to pay.

· Measured Service: Resource optimization and control capabilities are measurable, while measurement reports are transparent to the user.

Using the NIST cloud definition

As I mentioned at the beginning of this article, the industry has a different voice for how to define cloud computing. Individuals or businesses have a huge amount of investment in a cloud computing system, and the hype is also happening. As a cloud solution salesperson, you should be confident about cloud technology, and use simple language to articulate the deployment patterns and service patterns that the solution uses, preferably based on the cloud definitions given by NIST.

As an enterprise, you may be considering how to avoid cloud services that do not conform to the corporate objective of the deployment or service patterns, or you do not want to choose a cloud service solution that lacks one or several core features. Cloud solutions that are able to define tests through NIST generally have a good price/performance ratio. The "common features" (not the core definitions of NIST) are listed in the bottom column of Figure A, which can help you differentiate which cloud service functions are required by your business.

(Responsible editor: The good of the Legacy)