当前位置:网站首页>Merrill Lynch data technology expert team | building a cloud native product system based on containers
Merrill Lynch data technology expert team | building a cloud native product system based on containers
2022-06-26 09:15:00 【Merrill Lynch data tempodata】
Modern cloud applications often carry differentiated data services and complex application services , A stable and reliable cloud native product system is particularly important in the daily operation and maintenance of cloud applications . From boosting enterprises to realize digital transformation 、 From the perspective of promoting the innovative development of enterprise business , Put forward a set of “ Cloud native ” Architecture , The research contents of the architecture and the design of each layered structure are introduced in detail .
introduction
With the maturity of virtual technology and the popularization of related technologies of distributed applications , Most enterprises have gradually accepted the way of developing their own business based on cloud platform , The disadvantages of traditional hardware server deployment gradually show that it can not meet the needs of business expansion , In this context “ Cloud native ” The concept of... Came into being .
Matt Stine On 2013 First proposed in “ Cloud native ” Concept , And in 2015 Books published in 《 Move to cloud native architecture 》 This concept is further elaborated in . But up to now, different organizations and scholars have made great efforts to “ Cloud native ” There are still differences in the definition of , among CNCF( Cloud native Computing Foundation ) The definition of cloud nativity is : Cloud native technology is good for organizations in the public cloud 、 In new dynamic environments such as private and hybrid clouds , Build and run scalable applications . Think “ Cloud native ” It is an architectural idea based on cloud computing to solve the bottleneck of existing local deployment technology , It is the first thought for enterprises to realize digital transformation , Is to make full use of the existing cloud platform 、 Realize the rapid development of the project 、 deliver 、 A set of technical system and methodology for operation, maintenance and management .
An architecture based on the native products of the container cloud , The product business can be complicated in-house 、 Big data 、 More service 、 More cross departmental collaboration 、 The operation and maintenance cost is high , Help reduce the cost of the enterprise in the process of project deployment and operation and maintenance , Improve the efficiency of enterprises .
Cloud native system design
From the perspective of technical theory and architecture design, this paper expounds the “ Cloud native ” System and how it enables rapid deployment 、 Features and advantages of continuous delivery .
1、 System content
according to CNCF( Cloud native Computing Foundation ) Yes “ Native cloud ” The definition of and the current situation of R & D of the company's technology platform , From microservices 、API drive 、 Intelligent operation and maintenance and containerization are used to realize the construction of cloud native .
Microservice is a new system architecture idea in the evolution of software architecture , Microservices can solve single applications , It can realize the atomic deployment of applications , Simplify complex business , Divide and conquer can be realized among applications , And it can reduce the business coupling ;
Container is the carrier of microservice virtualization application deployment , You can apply it internally “ encapsulation ” To achieve mutual isolation between applications , Such isolation can effectively improve the overall agility and maintainability of the application , And it will be managed by scheduling to optimize resources , By applying “ loading ” To achieve mutual isolation between applications ;
Intelligent operation and maintenance It refers to the integration of human operation and maintenance knowledge 、 Experience and visual operation 、 Intelligent analysis 、 Combination of intelligent monitoring , Form a series of operation and maintenance decision-making theories and integrate them into the operation and maintenance system , So as to promote the frequent release of software projects in the production process 、 Rapid deployment 、 Fast validation reduces the cost of software delivery ;
API The driver implements the communication operation between services , It can effectively reduce the communication between applications 、 The complexity of deployment and operation and maintenance , It can effectively isolate the risks caused by directly operating the data of the underlying data model , It also provides a new way for the expansion of applications and services .
Cloud native content
2、 Architecture
About “ Cloud native ” The architecture of , From boosting enterprises to realize digital transformation 、 From the perspective of promoting the innovative development of enterprise business , Put forward “ Cloud native ” Architecture . This is the methodology for Merrill Lynch data to realize application cloud , The theory is based on the existing public cloud 、 Based on private cloud and hybrid cloud , With the concept of serving the enterprise, the application isolation operation is realized through the existing mainstream containerization technology , At the same time, it also integrates Merrill Lynch data The product system and related technologies involved in the application of visual operation and maintenance , It provides a convenient way to realize the visual management of cloud applications .
“ Cloud native ” Architecture diagram
“ Cloud native ” From top to bottom, the architecture can be divided into : Management services 、 Container service layer and platform service layer .
(1) Management services
The management service layer described in the article “ Cloud native ” The architecture design of is to provide a visual operation and maintenance platform for microservicing projects . In terms of configuration monitoring, the platform supports unified monitoring of applications 、 Unified configuration and other operations , In terms of deployment and upgrade, the platform realizes the independent deployment of applications 、 maintain 、 Expand 、 Upgrade and other operations . At the same time, the unified log center and zero intrusion link tracking are connected to the operation and maintenance and troubleshooting .
(2) Container service layer
The main purpose of the container service layer design is for the “ Cloud native ” The architecture provides high-performance scalable container application management capabilities , Support the whole life cycle management of enterprise containerized applications ; Help enterprises run the cloud efficiently Kubernetes Containerized applications .
The open source application container engine is used in the design process of the whole container architecture Docker To build microservices , And package the microservice dependencies together into a container , Then batch deploy to different environments ; Adopt open source container image warehouse Harbor To realize the construction of private warehouses within the enterprise ; Adopt container cluster management system Kubernetes To realize the automatic deployment of container clusters 、 Automatic volume expansion and shrinkage 、 Maintenance and other functions ; Open source enterprise container management platform is adopted Rancher To drive and adapt to different environments Kubernetes colony , It shields the user Kubernetes The underlying implementation details of , Provide uniform Rancher UI And Rancher API To realize the different Kubernetes The operation of deploying applications in the cluster .
(3) Platform service layer
The platform service layer mainly relies on the existing public cloud 、 Private cloud 、 Hybrid cloud and other new dynamically scalable environments to build and run elastically scalable applications , Based on virtualization technology 、 Provide infrastructure based on network , Integrate large-scale and scalable computing 、 Storage 、 data 、 The supercomputing mode that uses equally distributed computing resources to work cooperatively provides services for the container service layer .
Summary
What this article puts forward Merrill Lynch data “ Cloud native ” Architecture , It can be applied to the construction of Enterprise Cloud applications , Help enterprises realize the requirements of an integrated platform from internal resource integration to external resource management , Build an ecosystem integrating inside and outside , Enable unified provisioning 、 Centralized control 、 Intelligent monitoring 、 Flexible expansion . Based on providing a configuration visualization for enterprises 、 Delivery agility 、 The service extension is elastic and the cloud application is portable .
边栏推荐
- 教程1:Hello Behaviac
- Self taught neural network series - 1 Basic programming knowledge
- Applet realizes picture preloading (picture delayed loading)
- 《一周搞定模电》—功率放大器
- JSON file to XML file
- Application of hidden list menu and window transformation in selenium
- 行為樹XML文件 熱加載
- Bug encountered in training detectron2: the test set cannot be evaluated during training
- 小程序实现图片预加载(图片延迟加载)
- Pycharm occasionally encounters low disk space
猜你喜欢
随机推荐
Cookie session and token
Sqoop merge usage
Upgrade phpcms applet plug-in API interface to 4.3 (add batch acquisition interface, search interface, etc.)
phpcms小程序接口新增万能接口get_diy.php
Applet realizes picture preloading (picture delayed loading)
ImportError: ERROR: recursion is detected during loading of “cv2“ binary extensions. Check OpenCV in
[cloud primordial | kubernetes chapter] go deep into the foundation of all things - container (V)
Pytorch neural network
Dedecms applet plug-in is officially launched, and one click installation does not require any PHP or SQL Foundation
phpcms小程序插件4.0版正式上线
《一周搞定模电》—基本放大电路
Pycharm occasionally encounters low disk space
Phpcms V9 remove the phpsso module
修复小程序富文本组件不支持video视频封面、autoplay、controls等属性问题
Li Kou 399 [division evaluation] [joint query]
《一周搞定模电》—集成运算放大器
phpcms v9后台文章列表增加一键推送到百度功能
PD快充磁吸移动电源方案
How to compile builds
Nacos注册表结构和海量服务注册与并发读写原理 源码分析









