Shulou(Shulou.com)05/31 Report--

How to implement the landing practice of K8S container cloud CaaS platform, I believe many inexperienced people are at a loss about this. Therefore, this paper summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

Preface

In the era of mobile Internet, new technologies need new technology support environment, new software delivery processes and IT architecture, so as to achieve architecture platform, delivery continuity, and business services. Containers will become the standard deliverables for the new generation of applications, and the container cloud will help enterprise users build R & D processes and cloud platform infrastructure. Shorten the cycle of application delivery to the cloud and lower the operational threshold. Accelerate the dual transformation to Internet technology and business. The container cloud will dock with various code managed libraries to achieve automatic continuous integration and DOCKER image construction, laying the foundation for the integration of application delivery, development, operation and maintenance. The container cloud will support one-click deployment and delivery of applications, providing application lifecycle management services such as load balancing, private domain name binding, performance monitoring, and so on. It is an ideal platform for micro-service architecture, lightweight application deployment and operation and maintenance. In the future, the IT community will take containerized applications as the delivery standard. Container cloud provides a platform for developers and enterprises to quickly build, integrate, deploy and run containerized applications, so as to improve the iterative efficiency of application development, simplify operation and maintenance links, and reduce operation and maintenance costs.

For enterprises, the value of container cloud is reflected in:

As an internal technical link, multi-application unified deployment platform can improve development efficiency and reduce operation and maintenance costs.

In the process of platform production, the input is low and the output value is high. Can rely on the existing IaaS platform resources, compared with the traditional way to undertake project benefits, container cloud will accumulate valuable customer resources and data resources, these intangible assets will eventually win great strategic value for the company.

To increase the identification of cloud enterprises, if the cloud platform with only IaaS layer services is incomplete and the cloud services for users are insufficient, it should have a comprehensive cloud service system for users and an advanced and high value-added cloud platform.

Application functional architecture

The container cloud platform implemented based on Kubernetes can be divided into four parts in terms of function:

Automate the deployment, upgrade, and replication of containers

Expand or shrink the size of the container at any time to achieve elastic expansion of the container

Run, manage containers across machines in a clustered manner, and provide load balancing between containers

Kubernetes's self-healing mechanism makes the container cluster always run in the desired state of the user.

In the future, the container cloud may also become the ruler of the cloud management platform, and more applications will run natively on the container cloud. Because for the container cloud, it can abstract many scattered physical computing resources into a large resource pool, and use these resources to perform users' computing tasks. For users, operating a distributed cluster resource is like using a computer. With the centralized outbreak of artificial intelligence and the high requirements for fast iteration in the training process of machine learning, the container cloud, which can be accessed lightweight and can provide seconds of response to users, has a great advantage. The Google Tensorflow project with the same division also integrated with Kubernetes a long time ago, and Spark on Kubernetes and Hadoop on Kubernetes were implemented on the artificial intelligence platform. Both container technology, DevOps and micro-services show the gradual formation of a flexible, flexible and lightweight IT application model. The formation of this model has a positive significance for the rapid development of complex products. The contribution of Google and other IT giants to the industry has shortened the gap between start-ups and large IT enterprises and ushered in the best IT era. The container cloud architecture scheme has six functional modules, and each part contains the corresponding container technology stack. The specific structure is as follows

Storage scheme

The back-end storage is mainly driven by Ceph. Ceph uniquely provides object, block, and file storage capabilities in a unified system, which is reliable, easy to manage, and free software. Ceph is powerful enough to change your company's IT infrastructure and manage huge amounts of data. Ceph provides great scalability-allowing thousands of users to access PB and even EB-level data. Ceph nodes are supported by ordinary hardware and intelligent daemons. Ceph storage clusters organize a large number of nodes, which communicate with each other to replicate data and redistribute data dynamically. The main application scenario of Ceph in container cloud is stateful service. Business where data such as relational databases and nosql databases need to be persisted.

Network scheme

We initially used Calico in the underlying container network. Calico basic architecture diagram Calico is a pure three-layer network, without the introduction of a DP, no packets. Make another container inside the host, three can reach the terminal, you can know who has the problem, debugging is very easy, easy to manage. The application data in the container is completely isolated from the second layer. For most of our applications, only three layers are needed, and few applications deal with the second layer. And Calico supports rich network strategies and can achieve multi-tenant management, which is very important for the future provision of container cloud services.

Container arrangement scheme

Kubernetes is Google's open source container cluster management system, which is the open source version of Google's large-scale container management technology Borg for many years. The main features include:

1. Container-based application deployment, maintenance and rolling upgrade

two。 Load balancing and service discovery

3. Cluster scheduling across machines and regions

4. Automatic telescopic

5. Stateless service and stateful service

6. Extensive Volume support

7. Plug-in mechanism ensures scalability

Kubernetes has developed rapidly and has become a leader in container orchestration. Kubernetes provides a lot of features, which can simplify the workflow of applications and speed up development. In general, a successful application orchestration system requires strong automation, which is why Kubernetes is designed as an ecosystem platform for building components and tools to make it easier to deploy, extend, and manage applications. Users can use Label to organize and manage resources in their own way, and you can also use Annotation to customize the description of resources, such as providing status checks for management tools. In addition, the Kubernetes controller is built on the same API that developers and users use. Users can also write their own controllers and schedulers, and they can also extend the functions of the system through various plug-in mechanisms. This design makes it easy to build various application systems on top of Kubernetes. Currently, the entire kubernetes cluster is deployed with high availability, as shown in the following figure:

Performance monitoring scheme

Container monitoring objects mainly include Kubernetes clusters (components), application services, Pod, containers and networks, etc. These objects are mainly shown in the following three aspects: 1.Kubernetes cluster self-health monitoring (5 basic components, Docker, Etcd, Calico, etc.) 2. System performance monitoring, such as CPU, memory, disk, network, filesystem and processes, etc. Business resource status monitoring, including: rc/rs/deployment, Pod, Service, etc. For the overall operation status and performance status monitoring of the container, a self-developed monitoring system is used to achieve the unified monitoring of the overall IT resources.

Log collection scheme

The logging system of the container platform generally includes: the log of the Kubernetes component, the event log of the resource and the log of the application run by the container. This container cloud platform uses fluentd (launched by DaemonSet) to collect logs and send the collected logs to the unified log platform.

CI/CD scheme

CI/CD (continuous Integration and deployment) module shoulders the important task of DevOps and is the bridge between developers and operation and maintenance personnel. it realizes the automatic launch of business (application) from code to service, and meets the needs of one-click continuous integration and deployment in the development process. This container cloud platform connects with continuous integration & release system. In addition, the implementation of service scaling, HPA, load balancing, grayscale publishing, etc., as well as code quality check (Sonar), automated testing and performance testing plug-ins, these are important parts of the CI/CD PaaS platform.

Cloud platform function

Through the visual interface, the deployment, management and monitoring of resources can be realized simply and conveniently, including resource arrangement. The following functions can be achieved:

Comprehensive monitoring: hosts, routers, hard drives, public network IP, load balancers all have complete and comprehensive historical monitoring and real-time monitoring information.

two。 Open graphical operation: intuitively display the host and its related resources, and can directly operate the graphics, all changes can be updated automatically.

3. Operation log record: record all the important operation history of the user, easy to locate and find.

4. Network topology display: using graphics to show the complex network topology, connecting the private network and the host is more convenient and intuitive.

5. Backup chain visualization operation: the backup node is clear at a glance, at the same time, you can directly operate the new backup and rollback to the previous state.

6. Automatic scaling policy: define automatic scaling policy according to resource monitoring information, and adjust resource allocation or cluster size without human intervention.

7. Notification list: used to receive the execution results of monitoring alarm notifications, timer tasks or automatic scaling policies.

8. Rapid system construction: using resource choreography, you can quickly replicate an existing system with a complex topology in just a few minutes, or quickly plan your system architecture and evaluate the cost; it can also achieve cross-regional reuse of resource topologies.

9. Rich template creation methods: in the console, users can build templates from scratch, start with common templates recommended by the system, or continue to improve based on the templates already created; you can also extract a set of topological relationships from existing resources to become templates.

After reading the above, have you mastered the method of how to implement the landing practice of K8S container cloud CaaS platform? If you want to learn more skills or want to know more about it, you are welcome to follow the industry information channel, thank you for reading!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.