ZCCT_01_Introduction to Cloud Computing and Solutions
Hello, everyone. I'm mia. Welcome to zedstack, certified, cloud technician, training, course. Today, i would bring the first lesson. Introduction, to cloud computing, basic, technologies. And, solutions. Let's look at the objectives, of this course. Through this course, we can learn the following, knowledge. First is cloud computing, classification. And characteristics. Then our cloud computing, resource, pooling related, technologies. And their respective, problems, and virtualization. And cloud platform, solutions. Our course divided, into the following three points. Let us first move to the basic concept, of cloud. Computing. The most widely, accepted, definition, of the concept, of cloud computing, comes from the us national, institute, of standards, and technology. According, to the official nist definition. Cloud computing, is a model for enabling, ubiquitous. Convenient. On-demand, network, access to a shared pool of configurable, computing, resources. Like, networks. Servers. Storage. Applications. And services. That can be rapidly, provisioned, and released, with minimal management, effort or service provider, interaction. Suppose, we want to build a business system. Traditionally. First, we have to apply for the budget, then purchase, the hardware. Next waiting for the delivery. Then install the operating, system, and software. These processes, take weeks, or even months, and cost a lot of investment. And a waste bunch of time. Or you just need to visit the website, of the public, cloud service provider. Then you can select a cloud server, and has no difference, from the physical, server. Maybe only take a few minutes, and save a mass of energy, and investment. Therefore. We can list three values of cloud computing. Directly. It can significantly. Improve the efficiency, of our work, and it to reduce the cost of our it investment. Third, it can help us boost innovation. Because a lot of time is saved. Therefore. We can put more energy, into innovation. Let's take a look at the classification. Of cloud computing, interaction, methods. First is the private cloud. It is a private data center, and provides, internal, services. Its typical, characteristic. Is that it has reliable, security. Therefore. It is widely used in governments. Education. Medical, care, finance. Etc. The second is public, cloud, which provides, cloud services, for any network, users. Its typical, characteristic. Is that the initial, investment, is meager, and very flexible. Users can quickly expand, or reduce, their cloud services. Last is a hybrid, cloud. It combines, the private, cloud, and public cloud. For example. Users can run the database. In the private cloud, and run the web server in the public cloud. So it has the security. And compliance, of the private cloud. And the flexibility. And low cost of the public cloud. And it shows hybrid, cloud is a trend in the. Future. From the current perspective. Hybrid cloud is widely used in education. Media. Manufacture. Retail. And other industries. These are the three categories, of cloud delivery, models. There's several elements, to build an information, system, which are computer, room, necessary, hardware. Operation, system. Operation, environment. And business system. First, through infrastructure. As a service. The user does not need to care about the physical, environment. And operating, system. They are replaced, by ias, service. For example. When we buy a cloud host in a public service provider. We only have a host with an operating, system. And we have to build the business server by ourself. Then is the platform, as service. We can see the operating, environment. Is missing, which means we do not need to install the database.
Middleware. Or web server by ourselves, self, and they already delivered by pars. Sopars, is a higher cloud service than iaas. For example. Object, storage, is a standard, parse service. Last is the software, as service. It's very easy to understand. For the end user, there is no need to do anything. Just register, an account, then they can use the business system, and do not need to make other it investment. Like salesforce. Above all, they are the classification. Of cloud computing, service models. It can be understanding. That self-built, parts become cloud service, by different, layers. Next, let's move into resource, pooling technology. Cloud, computing, includes, the following, three very critical, technologies. The first one is server virtualization. Technology. The second network, virtualization. Technology. Next is distributed. Storage. Let's take a look at server virtualization. Technology. The picture on the left is the most traditional, physical resource model. There is a physical, server, and the operating, system, installs, directly, on it whether it is windows, or linux system. Inside, the operating, system, we would deploy, all kinds of business. Services. In this scenario. One of the most common problems, is the utilization. Rate of physical, server resources. Is very low, generally, between 10. And 50. Another, issue is that if we allocate, each new service, to a single, server. Their server amount will increase rapidly. So it causes the third issue, which brings hug pressure in our computer, room, because we have to consider, the space of computer, cabinets. Power supply. And the temperature. However. Let's move to virtualization. Mode. In this mode we would like to deploy virtualization. Software. Like es11. Kvm. Zen, and, hyper-v. Virtualization. Software, can split the physical resource, to install, multiple, virtual machines. Which means we can run various, services, in a single physical, machine. Therefore. It solves the utilization. Rate issue, which increases, to 50 percent to 80 percent. After that, the number of servers, is greatly decreased. So the computer, room resource, issue is solved accordingly. And what is the difference, between the physical mode and virtualization. Mode. The operation. And maintenance, are quite complicated. In traditional, physical, method. Because the administrator. Has to operate, different hardware. And it also has low efficiency. And high investment, issues. But in virtualization. Mode, it decouples, the relationship. Between the device. And the operating, system. So, it can easily share the hardware resource. And let operation. And maintenance, much easier. Although the virtual machine, has no difference with the physical, machine.
It Has a little performance, loss, because it brings in a virtualization. Layer. Next, let's look at the end network, virtualization. The left picture, is a traditional, network, architecture. And it's very easy to understand. In this architecture. All data transmission. Between servers depends, on physical, network. Equipment. However. In a virtualized. Environment. The operating, system, is running in a virtualization. Environment. Virtual, network, devices, carry its data transmission. The virtual, network, architecture. Is more flexible, than the traditional, physical, network, architecture. And also more efficient. Through this picture, we can find that all data transmission, between virtual machines, is carried by the virtual, switch, and it can also be transmitted, by virtual reuters. So it greatly reduces, the dependence, on traditional, physical, equipment. These for devices, are the most common virtual, network, devices. First is virtual, switch, which mainly responsible, for connecting, different virtual machines, in the same layer 2 network, to process, data frame forwarding, between virtual machines. The first photo, shows that these three virtual machines, are connected, by one virtual, switch. In which they can exchange, the layer 2 frame by the virtual, switch. Then is a virtual, router, which mainly responsible, for connecting, different layer 3 networks. And routing data packets, between, different, networks. In the second picture, the two virtual, machines connect to different switches. And switches, connect to the same one virtual, router. If the virtual machines, need to communicate. With each other, they need to transmit, data through the virtual, router. Next is the virtual, firewall. Which mainly responsible, for the access control, of data transmission. Between, different layer 3 networks, to protect network, security. In the same picture. Below. If we do not want the vmware to access the port 80 of vmb. Then we should configure, the policy, of the virtual firewall, to achieve our network security, demands. The last virtual, network, equipment, is virtual, load balancing. Which mainly responsible, for distributing. External, access, requests, to different virtual, machines, based on the strategies. For example. There are two virtual machines, in the third picture. And they are all running web service, using port 80.. Then we can create a load balancing. And a virtual ip, address for these two vms. When users, access the virtual, ip address, the load balancing, can balance, distribute, the traffic to tvms. Let's move to distributed. Storage, technology. Before, learning distributed, storage, technology. We must understand, the sand storage. As the left picture demonstrated. Then it is a centralized. Storage, architecture. It typically, contains, two controllers. And several blocks disks. The controllers. Are designed, as redundant. Several disks, usually, provide data security. By raid technology. The sand storage, devices, are generally connected, by fiber switches. And using two fiber switches, to give redundant, connections. The sand storage, also has its pros and cons. The most significant. Advantage, of sand storage. Is that its performance. Is outstanding. In small, and medium sized scenes. And for users, its maintainability. Is also very strong. But sand storage, has limited, scalability. It's no way to scale horizontally. For example. Using sun storage, in a cloud environment. The best practice, is that each sun storage, array only can connect, 10 to 15 computing, nodes. In other words, sun storage, has a single point of failure risk. If an unexpected. Failure, occurred in the sand storage. Then the entire cloud business services, will interrupt.
On The right is distributed. Storage, technology. As the right picture, shows, unlike, sand storage. Which is a distributed. Structure. And composed, of an ordinary, server. The distributed. Storage, software, can unify, the disks, which distributed. On different servers, as a large storage, pool through the network. With the development, of cloud computing, technology. Distributed. Storage, technology. Also develops, rapidly. And become more, mature. Due to its inherently, distributed. Architecture. It is very stable. But because it is a new technology. There is a specific, learning threshold, for users. Especially, in maintainability. Another specificity. Of distributed. Storage, technology. Is that its scalability. Is excellent. So, it solves the problem that traditional, sand storage, cannot scale horizontally. Next, it corresponds, with the cloud, which has the elasticity. Characteristic. And it is the critical, factor that users accept, and adopt this technology. Meanwhile. It does not have a single point of failure, issue, which means if a single hardware, device, in the storage, pool fails. It does not affect the cloud business services. In a word, distributed. Storage, technology, solve the two key issues of sand storage. First is the scalability. Issue, and another one is a single point of failure, issue. However. It also brings the learning threshold, for some users. Because it is a new technology. Next, let's look at cloud computing, solutions. Including, virtualization. Solutions. And cloud platform, solutions. Let's first look at the virtualization. Solution. In the virtualization. Platform, solution. The physical, server would install, a virtualized. Operating, software, as a virtualized. Computing, node, and install management, client in another physical, server, or virtual, machine. And it can dock fc san storage, or distributed. Storage, as back end storage, according to the demand. Generally. It adopts, all redundant, link design between the computing, node and back end storage, to avoid single points of failure. Above all, these are the virtualization. Platform, solution, brief, and its general deployment, process. Next, let's talk about the key features of the virtualization. Platform. First is resource, pooling technology. For example. There are three competing, nodes in these pictures. Each computing, node has 40 core cpus. One 2 8g, memory. After resource, pooling by virtualization. Software. And considering, the consumption, of the virtualization. Layer, it can provide, 120. Core cpus, and, 384. G memory virtualized. Resources. And taker for cpus. And he memory virtual machine, configuration. As an example. It can run 30 units. Another key feature, is resource, scheduling. Like high availability. Hot migration. Dynamic, resource scheduling. Elastic, scaling. Etc. These functions, can let the virtual, machine to cross the boundary, between physical, machines. Scheduling, on demand. These pictures, show the working, principle, of high availability. Three virtual machines, are running in two physical, machines. Two on the left. And one on the right. When one of the competing, nodes fails, its virtual, machine, will automatically. Restart, on another computing, node. High availability. Is the most critical, feature, in virtualization.
Software. But it has some prerequisites. Which are the two computing, nodes need to be in the same cluster. And connect to shared storage. And the other computing, node must have sufficient, resources. In this feature, there is a very critical judgment, mechanism. Which is how to judge whether the computing, node fails, and makes this feature very stable, and reliable. Another, essential, function, is hot migrating. Which can migrate, virtual machines, from the computing, node to other computing, nodes without downtime. It must be distinguished. From the previous, scenario. The previous, scenario. Is that one computing, node fails. And the virtual machine, on the failed node automatically. Restarts. On the other node. In this scenario. Virtual machines, migrate, to other computing, nodes, but here is the computing, node that fails. There are also some prerequisites. For hot migration. Which requires, the computing, nodes in the same cluster. And have the same type of cpu. Shared storage. Network, configuration. And other consistency. Requirements. The key to the feature, is the network, bandwidth. Which requires, a 10 gigabit, or better network connection. Between computing, nodes to meet the effectiveness. On the other hand, the migration, process, will copy the data in memory, so the accuracy, of memory verification. Is another key technology, of hot migration. However. There are still have some problems, that the virtualization. Solution, fails to solve. For example. There is an international. Corporation. And it has three data centers, which in london. New york, and melbourne. Particularly. And using the same virtualization. Software. The it, support team cannot, uniformly, manage the virtualization. Resource, in one, platform. So it brings some issues for them, such as they have to do a lot of repetitive, operation. Hard to manage the configuration. And cost analysis. And so on. So we develop a cloud platform, for handling, these issues. Before talking about the cloud platform. Let's first look at the differences, between the cloud platform. And the virtualization. Software. What are the typical differences. The main difference from virtualization. Software. Cloud platform, not only can do the resource, pooling, but can also do unified, management. And operation, of resources. And self-service. In the right picture. It demonstrates. The differences. Which the virtualization. Software, can only do resource, pooling, and resource schedules, like high availability. Hot migration. Snapshots. Clones. Etc. But the cloud platform, solution, can achieve multi-regional. Management. Heterogeneous. Platform, management. Multi-tenant. Management. Service, catalog. Metering. And billing, etc. In a word, the virtualization. Software, solves the problem of resources, schedule. And the cloud platform, solves the problem of resources. And resource, management. There are some typical cloud platform, solution, scenarios. The first one is multi-tenant. Management. It can split cloud resources, into different tenants.
This Demand, is very common, in education. And the internet, industry. For example. Different development, department, has its development, environment. And do not disrupt, by another department. So, every department, needs an isolated. Network, and resources. At the same time, it also can calculate, the usage of different departments. Particularly. Another scenario. Is multi-region. Management. It was also mentioned in the previous, scene. For example. There is an international, corporation. With headquarter. In new york, a branch in london. Melbourne. And shanghai. First of all the headquarters. Can uniformly, manage the i.t, infrastructure. Of all branches. Second, it can standardize, the image, software. And release. The regional, administrators. Do not have the privilege, to upload, personalized. Images. Third, the headquarters. Can control the resource permissions, of all corporations. The branch, only has regional management, authority. When the branch administrator. Cannot solve the problem, the headquarter, can provide, second line technical, support. However. The customer's, demand, general, is much more complicated. Than this. The third scenario. Is the heterogeneous. Platform, management. Which manages, multiple, platforms, in the one platform. For example. A user has a zed stack, a vmware, platform. And some public cloud service. The user wants to use one cloud management, platform, uniformly, to manage all resources. They provide a standardized. Service for end users. In this case, the user needs to adopt heterogeneous. Platform, management. Which can control, all different cloud platforms, by api. And shield the differences. Then present standardized. Service catalogs, to end users. The main heterogeneous. Platform, management, difference, for a cloud platform. Is that it does not deliver, and manage resources. It only docks other resources, management, tools. Finally. We will review the content, we talked about today through eight questions. What are the characteristics. Of public. Private. And hybrid clouds. What is the difference between ias. Paris. And saras. What are the fundamentals. Of virtualization. And specific, problems, solved by it. What kind of traditional, network, devices, can be used in a virtualized. Environment. What issues does distributed. Storage, mainly solve on sand storage. Describe, two basic scenarios. That virtualization. Solutions. Describe, two basic virtualization. Features. What are the essential, differences, between cloud computing, solutions. And virtualization. Solutions. Describe, two basic, cloud platform, management, features. Pause the video, and try to answer. The answers will be given on the next page. The answers, are one, public, cloud provides, cloud service for any user by the internet. But private, only provides, internal, service. And hybrid, is a server across public cloud, and private, cloud. 2., iis, provides, basic, hardware, and operating, system, environment. Paris, provides, a middleware, environment. And shields the hardware, and os. Sars, provides, software, service, and shields the hardware. Os. And middleware. Three, the fundamentals. Of virtualization. Are resource, scheduling. And it solves the resources, utilization. Rate issue of the physical, server. 4., v-switch. V-router. And v-load balancing. 5, scalability. And single point of failure. Six, resource, pooling, and resource schedule. Seven, high availability. Hot migration. Dynamic, resource, scheduling. And elastic, scaling. Eight, virtualization. Solutions. Cannot manage the resource, with different regions, and tenant. Nine, multi-tenant. Management. Multi-region. Management. And heterogeneous. Platform, management. For more information, or support. Please join our international, support, group, and visit our website. En.stack.io. That's all for this lesson. Thank. You.
2020-07-26 03:45
