Hi everyone, welcome to the first chapter in our Tencent Cloud practitioner course.Introduction to cloud computing. At the end of this chapter, you will have a better understanding of the history of data centers, the features of cloud computing, cloud computing technologies, the Impact and market of cloud computing and Tencent Cloud. In this chapter we will cover five sections, the evolution of data centers, features of cloud computing, cloud computing technologies, the Impact and market of cloud computing and Tencent Cloud. This video will cover the first section, the evolution of data centers. Subsequent videos will cover the remaining four sections. Let's get started with section one, the evolution of data centers. In this video, we will cover self-built EDCs and IDCs, Rented and Hosted IDCs, and Cloud Computing, before ending with a comparison of the three types of data storage services. There are a couple of challenges that Enterprise Data Centers face.Enterprise Data Centers are difficult to build and involve complex Ops management. They are hard to scale and adjust, and involve long launch cycles. Enterprise data centers also involve high total cost of ownership with capital expenditures, operational expenditures and opportunity costs. There is also a very uncertain total value of ownership, which amounts to the business value and benefits from IT. Data centers have five different layers, from 0 to 4. Layer 0 is the housing layer, which consists of buildings, logistics, site security, and fire prevention. Layer 1 is the facility layer, which consists of the power supply, cooling, cabinets and cabling. Layer 2 is the IT infrastructure layer, which consists of servers, networks, storage, and informational security. Layer 3 consists of the data analysis layer and the application support layer. The data analysis layer consists of databases, data warehouses and BI systems. The application support layer consists of middleware, messaging, the integrated development environment and APis And SDKs. Layer 4 is the business application layer, which consists of various application systems, such as enterprise resource planning, customer relationship management, office automation systems and more. Data centers can be divided into four Tiers based on their reliability and security, Ops management capabilities and infrastructure availability.Availability itself is equal to promised service time minus downtime, over promised service time times 100%. Tencent Cloud data centres are all above Tier 3.On the table to the right, you can see the description, availability and annual downtime of each tier. Tier 1 features basic capacity and an availability of 99.671%, resulting in a maximum annual downtime of 28.8 hours. Tencent cloud is Tier 3 or above, featuring parallel maintenance.Because Tencent cloud is Tier 3 or above, Tencent clouds availability must be at least 99.982%, with a maximum annual downtime of 1.6 hours. This is the basic calculation for data centers. Let's move on to Tencents self-built Internet Data Centers or IDCs. Tencent does all the work when building IDCs.From planning and construction and deployment, to operations and business system development. Having to take on all this work results in high costs, which is one of the main challenges involved in self built IDCs. As you watch the rest of this video, you can think about what other challenges may be involved in self built IDCs. Okay, now let's look at hosted and rented IDCs.Hosted and rented IDCs are two types of the leasing of IDCs resources such as storage, servers and bandwidth. The advantages of hosted and rented IDCs include lower costs, faster IDCs launch, carrier grade reliability, standardization, and Ops management. There are two responsible entities, ISPs and Enterprises. The ISP provides facilities, bandwidth and power for hosted IDCs. While providing facilities bandwidth, power, hardware management and maintenance for rented IDCs. The enterprise provides hardware management, maintenance and business systems for hosted IDCs. While providing business systems for rented IDCs. Let's talk about cloud computing. What is cloud computing? Cloud computing is an on demand and pay as you go service. The origin of cloud computing can be traced back to Nicholas Carr's article called it doesn't matter, that was published in 2003. In his article, Carr predicted that more and more companies would fulfill their IT requirements simply by purchasing fee based web services from third parties, similar to the way they currently by electric power or telecommunications services. The concept of cloud computing was proposed in 2006 by then CEO of google, Eric Schmidt.Amazon web services was also founded in 2006, marking the birth of cloud computing. The National Institute of Standards and Technology defines cloud computing as a model for enabling ubiquitous, convenient and on demand network access, to a shared pool of configurable computing resources. Cloud computing has many features. Cloud computing is characterized by massive scale with a public cloud often having hundreds of thousands or even millions of servers, and a private cloud having hundreds to thousands of servers. It is also characterized by high reliability, with multi replica fault tolerance providing high reliability. Cloud computing features tenant isolation, with multiple tenants sharing the underlying hardware resources, but being logically isolated at the upper layers. It also features dynamic scaling.With dynamic scaling, helping clients cope with the growth of their applications and user bases. Another important feature of cloud computing is that it is an on demand service, as it provides a large pool of resources that clients can purchase on demand. Lastly, cloud computing features monitorable and measurable resources, and low costs. Cloud platforms offer features for monitoring and measuring resources, and users only pay for the resources they use and not for the entire infrastructure. You can think of cloud computing as a type of utility. Imagine an electric power development center with factories that build their own power generation facilities. There are also factories that rent power generation plants and devices which are partially rented. Power can also be provided by a utility company for on demand usage. The three options are also available in IT infrastructure development. Companies can choose to build their own enterprise data centers, lease hosted IDCs or by on demand computing capability from providers. Let's compare enterprise data centers, Internet data centers, and cloud computing. The rental scope of EDCs is layer zero, layer one and part of layer two. In cloud computing's case, clients will rent everything from the cloud. In other words, from layer zero through layer four, the overall cost for EDCs is very high, while the overall costs for traditional IDCs are moderate and low for cloud computing. EDCs have a long launch cycle, while traditional IDCs have a moderate launch cycle, and cloud computing has a very short launch cycle. Regarding Ops management, EDCs are very complicated. Traditional IDCs are moderately complicated, and cloud computing is simple. EDCs are very difficult to scale, while traditional IDCs are moderately difficult to scale. In contrast, cloud computing can scale elastically, a concept that we will discuss in future videos. Lastly, EDCs feature high independence and controllability, while traditional IDCs feature moderate independence and controllability. Cloud computing features the public cloud, which means shared tendency and the private cloud, which means exclusive tendency.
