Cloud Native is undoubtedly the hottest word in the field of Cloud computing in the past two years. With the popularity of open source technologies such as Kubernetes and Docker, Cloud Native technology has been promoted to the ground quickly. Cloud native technology has first been widely practiced in emerging Internet enterprises, but also gradually entered traditional industries, and has been further developed to help thousands of industries to achieve digital transformation.
In the era of cloud native, the new challenge facing enterprises and developers is how to fully adapt enterprise architecture to cloud infrastructure, maximize the features of cloud computing, and achieve flexible expansion, stability, reliability, cost reduction and efficiency improvement of cloud native infrastructure.
In the process of containerization, Qunar has adapted portal platform, middleware, OPS infrastructure and monitoring of various systems. During this process, many problems have been encountered and a lot of experience in avoiding pits has been accumulated.
The cloud native technology directly hits every pain point in the express industry. The huge business volume of ZTO Express, fluctuation of traffic during peak periods, complex terminal types and business forms are all problems that the cloud native platform is good at solving, which makes the practical experience of Zto container team especially valuable.
More and more traditional services are migrated from physical machines and virtual machines to Kubernetes. However, most of the services deployed based on Kubernetes are stateless at present. What are the difficulties in realizing stateful service container? During the implementation of stateful service in K8s, New Oriental developed high-performance storage service XLSS to solve the problem of insufficient stateful service capability of native K8s.
After the implementation of containerization, CI/CD and other technologies, how to use open source tools and Kubernetes to design efficient management development test environment? And how to use IM system open platform to build chatops workflow? Hong Ya technology to give their actual combat experience.
As a global technology giant, Intel not only provides a wealth of cloud native application cases and solutions, but also has in-depth research on cloud native technology evolution, market trends, industry prospects and other aspects.
On July 29, CIC Cloud Computing Summit “Cloud Native Technology Practice Sub-Forum”, gathered from Intel, Qunar, ZTO, New Oriental, Hongya Technology technology, bring you real enterprise cloud biological, container practical experience.
Here are the highlights
Intel technology helps promote the implementation of powerful, easy-to-use, optimized and efficient cloud native infrastructure. The cloud native platform provides comprehensive acceleration for data processing, analysis and AI, and helps customers achieve flexible expansion, stability, reliability, cost reduction and efficiency increase of cloud native infrastructure.
Speech outline:
1. Cloud native development trend
Trend of digital transformation
Cloud native technology evolution and market trend
Cloud native industry picture
2, Intel power cloud native landing practice
Intel hardware and software technologies enable the cloud native architecture
Computing, storage, network and other technical analysis
3. Cloud native application cases and solutions
In recent years, container technology is very popular, and increasingly mature, many enterprises slowly began to container construction, and continue to explore and practice in the direction of cloud native technology. Based on this general trend, Qunar also took its first step towards cloud native – containerization at the end of 2020. As of early June 2021, qunar’s production environment has been connected to more than 150 applications, and the rest are being connected in succession. Looking back on the whole containerization landing process, there are actually many technical difficulties.
This sharing will mainly introduce the adaptation and transformation of qunar’s infrastructure, supporting tools for operation and maintenance, and publishing system in the process of container implementation on the cloud native road, as well as the problems encountered in the promotion process and solutions.
Speech outline:
1. The development history of containerization in Qunar
2. Qunar container practice
Plan to introduce
CI/CD platform transformation
Infrastructure Transformation
Adaptation of o&M tools
3, container landing process stepped on the pit
4, Roadmap
You will get:
Understand the difficulties of large-scale enterprise container landing process and system adaptation transformation scheme, more than a reference idea.
In 2019, ZTO Express will gradually switch from virtual machine to container-based scenarios, so as to realize operation and maintenance automation, rapid deployment of applications, elastic scaling and dynamic adjustment of application environment and resources, and improve the efficiency of R&D and operation. After satisfying the business of the company, use cloud native technology to adapt to more business and scenarios.
This time, we will share the practical experience and architecture decision-making ideas of ZTO container team: K8s multi-cluster planning, edge cloud computing, Service Mesh practice, etc.
Speech outline:
1. Current situation of business system architecture and internal drive of cloud infrastructure
Bottlenecks and methods of microservice architecture upgrade
2, the application of portrait, resource measurement
3. Business scenario requirement analysis of multi-cluster management
Benefits and risks of federated clustering in multiple data centers
Design practical multi-cluster management according to different SLAs
4. Edge cloud: Natural advantages and relative disadvantages of containers in edge computing
5. Problems encountered in the application and practice of Istio in ZHONGtong business
You will get:
Through this share, want to in production deployment and implementation team K8s cloud native environment, can learn K8s community since 2019 technology components in a production environment of actual usage, and how to from each kind of predominantly virtual machine to upgrade to the data center is given priority to with container path, at the same time also can learn in the process of landing, Application of cutting-edge technologies such as Istio and Service Mesh to actual scenarios. Learn how to make architecture trade-offs based on the enterprise’s own technology level.
Following the general trend of cloud computing development, New Oriental continuously promotes the construction of cloud service. As more and more applications migrate to containers, migrating stateful services such as Kafka and Redis is a challenge that requires performance without sacrificing application availability. In the process of cloud transformation of stateful service, New Oriental adopts the mode of combining customized Operator with self-developed local storage service, which enhances the capability of K8s native local storage solution and steadily pushes forward the containerization construction of the enterprise.
This sharing will mainly introduce new Oriental stateful service construction process thinking and self-developed local storage service implementation ideas.
Speech outline:
1. Status quo of New Oriental stateful service in K8s
2. Ability analysis of native K8s to support stateful service
3. Pain points solved by self-developed storage product XLSS and the implementation ideas behind
4. Implementation of XLSS: large-scale storage middleware service
5. Manage K8s platform based on KubeSphere
6. Stateful services container RoadMap
You will get:
Understand the challenges, thinking and practice of New Oriental in the implementation of stateful service in K8s. In order to solve the problem that native K8s cannot support stateful service, the implementation ideas and implementation process behind New Oriental’s self-developed high-performance storage service XLSS.
With the implementation and promotion of cloud native, container-based and micro-service, the number of components in the application becomes more and more, and the logic of the organization becomes more and more complex. In the process of development and testing, project coordination and frequent switching of multiple consoles of CI/CD affect the improvement of efficiency.
This sharing will introduce hongya technology development test environment management and chatOPS practice experience based on CI/CD, project system and IM tools.
Speech outline:
1. CI/CD landing status
Application definition selection
Deployment tool Selection
Quickly build development test environment based on CD
2. Build chatops tools based on CI/CD and IM open platforms
Current problems and solutions
Develop test process models, iteration-grained groups, message notifications, iteration-grained test environments
3. Rapid debugging and development of test environment
Zero delay debugging in development test environment
You will get:
After the implementation of containerization, CI/CD and other technologies, how to use open source tools and Kubernetes to design an efficient management and development test environment. And how to use IM system open platform to build chatops workflow.