当前位置:网站首页>[teacher Zhao Yuqiang] Alibaba cloud big data ACP certified Alibaba big data product system
[teacher Zhao Yuqiang] Alibaba cloud big data ACP certified Alibaba big data product system
2022-07-03 05:46:00 【Teacher zhaoyuqiang】
Alibaba's big data product system is based on the data processing services of Alibaba cloud Apsaras platform . It is mainly divided into Alibaba cloud big data infrastructure products and Alibaba cloud digital plus platform , The product architecture is shown below :
One 、 Alibaba cloud big data infrastructure products
1、 Cloud database ——RDS(ApsaraDB for RDS For short )
- Stable and reliable 、 Scalable online database services
- Open and use ,DMS Visual interface
- compatible MySQL,SQL server,PG Relational database
- Provide database online expansion , Backup rollback , Performance monitoring and analysis functions
- Read only instances and temporary instances
advantage :
- hot standby —— Second switching , Service availability reaches 99.5%
- Safety protection —— prevent DDOS attack ,SQL Injection warning , Multiple backup of data
- Simple and easy to use —— One click data migration , Visual management operations
2、 Distributed relational database ——DRDS(Distributed Relational Database Service)
- be based on RDS Distributed data storage and retrieval products
- Horizontal split can be expanded smoothly
- Solve user list RDS Unable to support the suffering of the business
- Reduce the difficulty of using distributed database
advantage :
- Simple and easy to use —— compatible MySQL( Interaction protocol 、SQL)
- Stable and reliable —— Share Alibaba components
- Distributed —— Horizontal split , The capacity is 100 times that of a single node
- Scalable —— Increasing or decreasing nodes has little effect on the application , Efficient data migration
3、 Form storage ——TS(Table Store)
- Built on Alibaba cloud Apsaras distributed system NoSQL Data storage services
- Massive structured data storage and real-time access
- Flexible resource reservation
- Real time monitoring display
advantage :
- Stable —— Automatic fault detection and recovery , System availability 99.9%
- Security —— User level data isolation 、 Access control and rights management , Data redundancy backup
- On a large scale —— From a single watch to a hundred TB Level data storage
- High performance —— Millisecond level single line read and write latency , One hundred thousand QPS
4、 Analytical database ——ADB(Analytic DB)
- Cloud computing platform for real-time and high concurrency online analysis of massive data
- Free computing and query capabilities
- High availability and high security
- Full compatibility MySQL agreement
advantage :
- A high degree of computational freedom —— adopt SQL Flexible multidimensional analysis 、 Pivoting 、 Data filtering, etc
- Rapid response time —— Millisecond level of 100 billion level data perspective , Millisecond level large table Association computation
- Easy to use —— standard SQL, Support standards MySQL agreement , Built in a variety of cloud platform data input and output
- Rich features and functions —— High performance automatic indexing , The rapid export of massive data, etc
5、 Big data computing services ——MaxCompute
- in the light of TP/BP Level data 、 Real time requires less distributed processing power
- Big data computing power
- Open the box
- Data security
advantage :
- Distributed —— Distributed cluster architecture , Flexible extension
- Security —— Automatic storage fault tolerance mechanism , All the calculations are done in the sandbox
- Ease of use —— Full support based on SQL Data processing of , Provide standards API, Data upload and download with high concurrency and high throughput
- Management and Authorization —— Multi user management and collaborative data analysis , There are many ways to manage user rights , Flexible data access control decisions
6、 Data integration (Data Integration)
Data integration is a stable and efficient service provided by Alibaba group 、 Elastic and scalable data synchronization platform , Offline services for alicloud big data computing engine ( Batch ) Data access .
advantage :
- many : There are many kinds of data sources , Multiple data channels , Complete data transmission mode , Rich data processing plug-ins ;
- fast : Efficient way to call , Strong transmission speed , Powerful throughput ;
- good : Robust transmission channels , Intelligent error detection , Automatic transmission recovery ;
- province : Open the box , Dynamic allocation , Stretch , Apply on demand , Pay as you go ;
7、 Object storage (Object Storage Service, abbreviation OSS)
- For amounts 、 Security 、 Low cost 、 Highly reliable cloud storage services ;
- Open and use , Storage clusters with infinite space ;
- adopt API/SDK Interface or OSS Migration tools are convenient to move massive data in or out ;
- Storage object operations are atomic , Strong consistency ;
advantage :
- reliable : Service availability 99.99%, Data persistence 99.999999999%(9 individual 9), Multiple backups , Scale automatically ;
- Security : User level resource isolation , Disaster tolerance in other places , Enterprise level multi-layer security protection , Multiple authorization mechanisms ;
- Low cost : Multi line GBP Backbone network , No bandwidth limitation , Upstream traffic is free ;
- Multiple types of data processing capabilities : The image processing 、 Audio and video transcoding 、 Content accelerated distribution 、 Yellow service , Archiving services, etc .
Two 、 Alibaba cloud digital plus platform
Alicloud's data products are integrated into the data plus platform , The positioning of alicloud's public cloud plus platform : One stop data platform ( Integration includes from basic data analysis and application to big data development 、 Dispatch 、 Operation and maintenance , To machine learning and so on ); Three layers of service ( Underlying computing 、 Data platform analysis tools 、 Application layer service ) Industry solutions .
1、DataWorks( primary Data IDE)
Data workshop DataWorks( The original big data development kit Data IDE) Is based on MaxCompute As a computing and storage engine, massive data offline processing and analysis platform for workflow visualization development and hosting scheduling operation and maintenance .
advantage :
- major : Ali for many years DW/BI Experience precipitates , Full link solution , High efficiency and low cost ;
- Powerful : Integrated component services , Multiple heterogeneous data sources support , Multi person collaborative code development , Perfect version management , minute 、 Hourly scheduling 、 Drag and drop data analysis and visualization algorithm modeling ;
- Big data processing capabilities : Perfect fusion Max Compute, Support the orderly operation and management of 100000 level tasks .
2、Quick BI
Provide massive data real-time online analysis services , Supports drag and drop operation , It provides rich visualization effects , It's easy to do data analysis , Business data exploration , Report making, etc .
advantage :
- Low threshold : Drag and drop , Simple and easy to use ;
- Strong function : A variety of solutions , Rich means of presentation ;
- Big data processing capabilities : Data analysis , Powerful data processing capabilities
3、 machine learning PAI
- be based on MaxCompute、GPU colony , Support MR、MPI、SQL、BSP、SPARK And so on ;
- Built in Ali 、 Distributed algorithm of ant precipitation , Support 10 billion level data training ;
- WEB Interface , By dragging 、 PULL 、 Drag and other ways to complete the complex data mining process ;
advantage :
- Provide a one-stop platform service from data preprocessing to model evaluation , Significantly reduce the threshold of big data algorithm modeling ;
- Support custom algorithms and components , Flexible and open personalization , It greatly improves the modeling efficiency ;
- Provide rich distributed algorithms , Improve model accuracy , Help mining business value from massive data .
边栏推荐
- 2022.7.2 simulation match
- Transferring images using flask
- chromedriver对应版本下载
- [untitled]
- Ensemble, série shuishu] jour 9
- 期末复习(day3)
- Progressive multi grasp detection using grasp path for rgbd images
- ES 2022 正式发布!有哪些新特性?
- 2022.7.2day594
- "C and pointer" - Chapter 13 function of function pointer 1 - callback function 1
猜你喜欢
期末复习(Day5)
Method of finding prime number
【一起上水硕系列】Day 10
[explain in depth the creation and destruction of function stack frames] | detailed analysis + graphic analysis
[together Shangshui Shuo series] day 7 content +day8
Beaucoup de CTO ont été tués aujourd'hui parce qu'il n'a pas fait d'affaires
[advanced pointer (1)] | detailed explanation of character pointer, pointer array, array pointer
Analysis of the example of network subnet division in secondary vocational school
DEX net 2.0 for crawl detection
Redis使用Lua脚本简介
随机推荐
QT read write excel -- qxlsx insert chart 5
C 语言文件操作函数大全 (超详细)
6.23星期四库作业
JS implements the problem of closing the current child window and refreshing the parent window
Calculation method of AUC
2022.6.30DAY591
期末复习(Day5)
kubernetes资源对象介绍及常用命令(五)-(ConfigMap)
Ansible firewall firewalld setting
How do I migrate my altaro VM backup configuration to another machine?
70 shell script interview questions and answers
Final review (day3)
AtCoder Beginner Contest 258(A-D)
CAD插件的安装和自动加载dll、arx
獲取並監控遠程服務器日志
Beaucoup de CTO ont été tués aujourd'hui parce qu'il n'a pas fait d'affaires
Personal outlook | looking forward to the future from Xiaobai's self analysis and future planning
Making coco datasets
CAD插件的安裝和自動加載dll、arx
[minesweeping of two-dimensional array application] | [simple version] [detailed steps + code]