For small and medium enterprises, the IT infrastructure within the company must always be “up to date” in order to grow the business, often providing more resources than are currently required. Therefore, a fully functional data center is essential for most companies. However, it is expensive to set up your own data center and hire professionals to maintain it, as well as difficult to scale and adjust it flexibly on demand. In response to this demand, cloud computing came into being.

With the help of cloud computing, application services, including storage space, processing capacity, database and software environment, can be directly used on the cloud through the Internet without buying hardware servers.

The history of cloud computing

This type of networked computer, which gives users more processing power and storage, is not a new invention. The first steps towards cloud computing were taken in the 1950s with the mainframe computer. At that time, users could access the mainframe and use its functionality through multiple other terminals in the organization, although users had to time the mainframe well to use it.

In the decades that followed, virtualization emerged to allow an abstract reconstruction of computer instances. The virtual environment was then made available to all network users and commercially available to many more in the 1990s. The concept of the cloud became more and more popular, and by the year 2000, many companies were becoming interested in the cloud.

Today, most smartphones (or the Internet of Things) are connected to the cloud all the time, and cloud computing has not only brought convenience to companies, but also become a part of everyday life for many people. So what is cloud computing?

Cloud computing is a general term for hardware and software services provided through the Internet. It usually centers on large data centers or server clusters, pooling resources of multiple computers or servers together to form grid computing to achieve high performance. It covers everything from cloud storage space provided by remote servers to infrastructure in the cloud, and when combined with virtualization, can create a single virtual instance for a single user in the network. Although cloud computing involves a lot of things, there are clear rules about whether it can be called cloud computing or not.

According to the NIST paper, a service must satisfy the following characteristics to be called cloud computing:

  • Self-service on demand: Users should always be able to request the required resources independently, without having to contact the service provider.
  • Extensive network access: Access cloud computing anytime, anywhere with any network device.
  • Multi-person Shared Resource Pool: Pooling multiple computing resources is a basic requirement of cloud computing. The resource pool can dynamically allocate and reallocate resources, such as processing power and storage, to meet the needs of multiple users.
  • Rapid redeployment flexibility: Resource delivery must be fast and requirement-based. It can be automatically zoomed in and out in some cases without the intervention of the customer or service provider.
  • Measurable services: Always monitor the usage of cloud services, increasing transparency for both parties.

Various types of cloud computing

With the increasing number of cloud services today, there are various cloud solutions on the market. In addition to the price and support provided, the main differences between these solutions are the layers and deployment models provided. The “layer” refers to the level of services, and the deployment model determines the types of services provided.

The service level

The “tier” refers to the corresponding specific service level, and the scope of the service is generally described through various “service as a service” levels. Therefore, a pyramid model is often used to present them. Infrastructure as a Service has the greatest scope, while Software as a Service focuses only on specific applications.

The specific coverage of these “Service as a Service” is as follows:

  • Infrastructure as a Service (IaaS) : At this level, vendors offer complete hardware solutions: computing power, storage space, and networking. Instances used by users are completely virtual and are pooled. IaaS can be used as a basis for other layers or can be provided as a stand-alone product.
  • Platform as a Service (PaaS) : Provide a complete cloud environment with more than just hardware. PaaS is primarily aimed at software developers. The service provider provides an established development environment in a cloud hosted on hardware. As a result, programmers can save time and expense in maintaining such an environment.
  • Software as a Service (SaaS) : Provides software solutions for users. SaaS is primarily for the typical end user, who doesn’t have to worry about installing and maintaining software, and can rest assured that the hardware is powerful enough.
  • Everything is a Service (XAAS) : XAAS can be categorized at different levels. Because in addition to the three layers mentioned above, there are special reasons (such as marketing) that individual providers offer additional services like XAAS. This is usually for special reasons (such as marketing). In fact, XAAS can be classified into different levels.
  • People as a Service (HuaaS) : A form of crowdsourcing in which a group of people do their work over the Internet.

Of course, the three main service models we see on a daily basis are IaaS, PaaS, and SaaS.

Deployment model

The deployment model describes four types of services:

  • Private cloud: The server is used by only one customer. Private clouds can be local (internal clouds). Even if a hosted program for a server cluster is used, other customers using that program cannot access the dedicated hardware.
  • Public cloud: Public cloud is considered as the main form of cloud computing. Its core attribute is shared resource service, that is, everyone shares server resources together. As users, we cannot see who is using server resources and changes.
  • Hybrid cloud: This model is a mixture of private and public clouds. This means that users can determine which parts of their operations to keep private (for example, security aspects) and which parts to keep public.
  • Community cloud: A community cloud functions like a private cloud, except that multiple users share a dedicated hardware instance. However, the user resource pool is not random and is usually composed of customers from the same business area or with similar interests. In addition, community clouds can be managed either internally or externally. More resource savings than running multiple private clouds.

The pros and cons of cloud computing

The emergence of cloud computing provides a solution to the problem of high cost for small and medium-sized enterprises to build their own IT infrastructure. The use and maintenance of the hardware is taken over by the service provider, and the enterprise only needs to invest in cheap terminals to access the cloud computing services. In addition, professional server clusters have much higher security than most organizations manage on their own, with dedicated security personnel and server specialists protecting the data center at the physical level. In addition, most cloud computing service providers have mandatory backup of all data, which can effectively avoid the hidden danger caused by data loss.

As a relatively new service feature, cloud computing has some drawbacks. For example, depending on the service provider, the level of customizability in terms of configuration may vary greatly and may not meet the needs of a particular enterprise. At the same time, the use of cloud services will be too dependent on the provider, once they encounter technical problems, will directly affect the business operation of the enterprise. The use of cloud computing also depends on a strong and stable Internet connection, otherwise enterprise employees will not be able to work efficiently.

Despite the above disadvantages, the biggest concern for cloud solutions is data privacy. While cloud providers can secure stored data, transferring data over the Internet always poses a security risk.

Cloud computing, after all, is not free, and organizations must carefully consider the time required to use resource capacity in order to maximize cost savings. In this respect, and a cloud in the scene, the CDN of core business, to provide cloud storage, cloud, cloud security, traffic marketing, such as cloud services, to help users realize content delivery speed, acceleration, business growth, product research and development, and on-demand billing, WebP, h. 265, rate adaptive speed limits and other cost control related functions, Helping businesses save money.

Recommended reading

Say goodbye to DNS hijack, read DOH

[vernacular popular science] chat about those little knowledge of DNS