On September 16th, Distributed Cloud | 2021 Global Distributed Cloud Conference Shanghai Station was grandly held. Under the unremitting evangelization of the Global Distributed Cloud Conference, the cloud computing industry has paid more and more attention to distributed cloud, represented by the members of the Global Distributed Cloud Alliance, a large number of distributed cloud technologies and practical achievements have emerged, laying a solid foundation for the development of distributed cloud computing.
The 2021 Global Distributed Cloud Conference adds another powerful impetus to the development of distributed cloud computing, and this conference has four major forums: distributed cloud theme report, edge cloud forum, cloud native thematic forum, and distributed database forum, focusing on distributed cloud, edge computing power, cloud native, distributed architecture and other technologies and practices. The Global Distributed Cloud Alliance, together with Alibaba Cloud, Tencent Cloud, Google Cloud, ZTE, JD Cloud, Ammann Cloud, Netxin Technology and other top distributed cloud technology service providers at home and abroad, talked about the new trend of distributed cloud innovation, sought a new future for cloud computing reform, and shared the new dividends of distributed cloud computing!
At the cloud-native forum held on the afternoon of September 16, Zhang Yang, CEO of Seconds Cloud, delivered a keynote speech entitled "O&M Trends in the Cloud-Native Era". He said that the future trend of container management must be more marginal and more decentralized multiple clusters, rather than a large, centralized cluster. What The second cloud has to do is to help every enterprise easily manage the cloud native environment, so that everyone can easily become a cloud native O&M master!
Cloud-native reinvents the IT technology landscape
Zhang Yang said that the cloud computing industry has undergone a long evolution from the physical machine era, to the virtualization era, and then to the cloud native era. From bare metal to the era of virtualization, the use of the entire architecture and software and software design have not changed much, although microservices exist in the era of virtualization, but the large-scale application of microservices is in the container era. Cloud native is one of the biggest trends in the software industry today, and has changed the way the industry thinks about developing, deploying, and operating software products!
Zhang Yang said that the transformation from a single architecture to a microservice architecture, including extreme auto scaling and self-healing of failures, will definitely bring new changes and great changes in the cloud-native era. And in the process of change, kubernetes must be mentioned. At present, Kubernetes has become the norm for enterprise deployments, with 82% of enterprises in China already using Kubernetes. Each enterprise R & D department, testing department will establish a different Kubernetes to meet the needs of the department. However, Kubernetes itself has complex operations and lacks enterprise-level functions, and Kubernetes mainly implements container scheduling and orchestration, if you need to implement container monitoring, operation and maintenance, including CI/CD, etc., you need about 30+ open source components to run the complete "cloud native environment". How to help enterprises achieve unified management, unified security control, and how to reduce O&M difficulties and reduce O&M personnel costs for enterprises, Second Cloud provides enterprises with an out-of-the-box software to solve the management, operation and maintenance and construction problems of enterprise cloud native system in one stop!
Zhang Yang said that like some factories, it is necessary to establish an IT environment in each plant, and they are a cloud-native environment at the edge. And many factories are located in the city can not find the operation and maintenance personnel who understand Kubernetes, the value of the second cloud is reflected in this, the second cloud can not only support the unified management and operation of multiple Kubernetes in the center, but also allow the edge operation and maintenance personnel to quickly get started.
The trend of O&M in the cloud-native era
Zhang Yang believes that the traditional monitoring can not adapt to the cloud native environment, compared with the traditional failure collection unified view, in the era of containers, Kubernetes, the "human flesh mode" can not be determined, only the use of AI algorithm capabilities for machine learning, alarm merger, classification and analysis, so as to quickly give fault judgment.
At the same time, in the era of containers and Kubernetes, the traditional application microservice is split into several/dozens of containers, each of which generates logs, and the amount of logs increases several times. As a complement to monitoring, the concept of observability is introduced to lower the threshold of experience and uncertainty. Observability in a very low threshold and image way to allow operators to visualize, clearly see the current system status, Zhang Yang believes that observability is to visually tell the data to customers. For example, if a person's body is sick, medically you can mark this piece in red. The system is the same, once the system is red, the operation and maintenance personnel have no experience and know that the red is a failure. The changes brought about by observability make accurate judgments that operators do not need to accumulate a lot of operation and maintenance experience.
Cloud-native intelligent O&M middle office
Zhang Yang said that in the cloud-native era of O&M, it is urgent to connect traditional O&M data and cloud-native O&M data, so that customers can have full-link, full-stack system O&M management capabilities on one page to achieve "sensitive and stable" O&M.
Zhang Yang said that the second cloud product design concept mainly helps customers solve how to better and more intelligently solve operation and maintenance problems in the cloud-native era, including declarative APIs, pluggable functions, comprehensive observability, ease of use, service governance, AIOps and even NoOps.
(Cloud-native intelligent O&M middle office architecture diagram)
Zhang Yang introduced that the gray part of the underlying layer of the architecture diagram is the basic capability, based on this, customers can choose to deploy Kubernetes, Rancher and other environments, and use the second cloud to achieve management. Seconds Cloud can also provide customers with the ability to deploy K8s with one click, helping customers deploy a complete, highly available, cloud-native environment with zero code visualization in one click. Customers have an algorithm middle platform in O&M, and intelligently analyze all data from multiple algorithms and sources. Not only that, the second cloud can also help customers do business-level data display and analysis.
In the visualized middle platform, the content that customers want to show can be flexibly configured and visualized, and all management behaviors and multi-tenancy are uniformly configured through a middle platform to form a display of various capabilities.
Out-of-the-box container cloud platform
Zhang Yang said that first of all, the second cloud will provide customers with an out-of-the-box container cloud platform, including the previous 30+ components, which are all one-click deployment, at the same time, customers can also do plug-in deployment according to their needs. Secondly, Second Cloud will provide a concise and unified visualization page, so that customers have a consistent interface operation experience. Third, On the basis of open source, Seconds Cloud has done a lot of security compliance enhancement and optimization checks, including configuration and manual repair. Fourth, in terms of application migration, Seconds Cloud supports the creation of virtual machines for applications that cannot be containerized temporarily, thereby reducing the cost of traditional application migration. Finally, in terms of production-oriented, Seconds Cloud has features such as multi-network support, data permission management, resource quota management, data consistency, and online hot upgrade.
Unified orchestration and scheduling of multiple Kubernetes
Zhang Yang said that multi-Kubernetes include several categories, different Kubernetes publishers, different Kubernetes versions, different hardware platforms and so on. Entering the era of containerization, the second cloud can realize the mixing of X86 servers and multi-domestic servers. In terms of xinchuang, second cloud has also done a full range of support:
At the processor level, Second Cloud supports domestic processors of different architectures such as Haiguang, Mega core, Feiteng, Kunpeng, etc., and is currently the only container cloud platform in China that supports Shenwei processors;
At the operating system level, it supports domestic operating systems such as Winning Bidding Kirin, Galaxy Kirin, Tongxin UOS, and Deepin System;
At the server hardware level, it supports domestic servers such as Huawei Taishan, Zhongke Shuguang, Zhongke Controllable, Great Wall, and Huacheng Jinrui.
For multiple cross-region, cross-room, and multi-cloud Kubernetes clusters, The Second Cloud can realize cross-cluster application release and automatic fault migration. SupportScopic deployment of O&M analytics components in multiple locations, sinking computing power, centralized/cloud analysis result aggregation, no need to upload massive O&M data, multi-level, multi-K8s cloud native observability and centralized O&M.
AIOps for data-driven response mechanisms
(AIOps diagram of data-driven response mechanisms)
Zhang Yang introduced that the data-driven AIOps of The Second Cloud can help customers collect all the data, help customers match according to the algorithm and data, and achieve the optimal algorithm model or algorithm scenario that best meets the customer, at the same time, the Second Cloud also cooperates with third parties to help customers achieve intelligent log anomaly detection, trend prediction, root cause positioning, etc.
Full-link cloud-native observability
Zhang Yang said that the solution to the problem of failure and operation problems in seconds cloud is called full-link cloud-native observability. The so-called full link refers to the migration from the underlying hardware server to the network security device to the cloud-native environment. Realize all-scenario observability, all-index monitorability, and global intelligent O&M. Zero code customized data analysis "APP" A set of platforms, one collection, one data, point-and-click, to meet various data analysis needs, fast fault location and one-click monitoring and diagnosis. Includes data analysis models, charts, alarms, panoramic views, interface menus... All can be customized on demand with zero code in a fully graphical manner. Professionals can also develop analytics plug-ins on the platform via coding (SPL/regular/scripting). Meet the personalized needs of thousands of people, and can meet the requirements of different departments and different authority personnel such as O&M/network security management/business management personnel, display different analysis results and pages, and meet the requirements of multi-source data comprehensive analysis.
Seconds Cloud helps enterprises easily solve operational challenges in the cloud-native era
Second Cloud is the first company in China to focus on cloud-native O&M, and in the era when the K8s ecosystem has become the normalization of enterprise cloud-native platforms, it helps enterprises to unify management, intelligent O&M, quickly build cloud-native systems, and improve the management, ease of use and observability of cloud-native systems. Designed for the O&M scenarios of enterprise cloud-native systems, Second Cloud is powerful, simple to use, has excellent experience, and quickly empowers traditional enterprise cloud-native upgrades and transformations.