当前位置:网站首页>Continuous definition of SaaS mode cloud data warehouse + bi

Continuous definition of SaaS mode cloud data warehouse + bi

2020-12-07 19:19:11 Aliyun yunqi

Cloud data warehouse Overview

Today, I'd like to talk about our Saas Cloud data warehouse plus business intelligence under the mode BI Something new can come out . Let's start with an overview of cloud data warehousing . Predict 2025 year , Global data growth to 175ZB, China's data volume has grown to 48.6ZB. Under the premise of data explosion , Let's see BI The growth of market size . Predict 2023 year , We China BI The compound annual growth rate of software market is 32%. Cloud computing is also growing ,2019 The growth rate of China's cloud data market has reached 66.9%.

image.png

Cloud data warehouse allows enterprises to create and start using data warehouse services in a few minutes , At a lower cost , Focus on business , Through the diversification of large-scale data processing 、 mining 、 analysis , Get business insight quickly . It has four characteristics : Large scale data analysis , High performance , Flexible expansion , Low cost .

image.png

BI Usage scenarios and Trends

business intelligence (BI,Business Intelligence) It is an information system established for the purpose of providing decision-making and analytical operation data . With the development of our society and the explosion of data , Supported by so much data , Enterprises hope to quickly mine some more scientific data from these data , Then we have a scientific and data-based decision-making help for our enterprises . meanwhile ,BI Will also help enterprises to use a refined operation , Customer relationship maintenance , And cost control .
Let's take a look at business intelligence building an information system, its main process . The first is data access , Integrate and integrate all kinds of data scattered inside and outside our enterprise . And then we go to a data preparation stage , It's just one. ETL The stage of . And then to a data analysis stage , Finally, these results will be handed over to the decision-making level , The decision-making level can make some decisions through this data . No matter it's fine operation , Or the customer maintains the relationship , Or cost control , We can get some help from these data .

image.png

As the amount of data skyrocketed , Our business is growing rapidly , A variety of analytical requirements have been created . It's not just about analyzing diversity , And want real-time , For example, second level instant query . At the same time, based on such a large amount of data , More and more attention has been paid to the security and compliance of data . So we need to quickly integrate multi system data and achieve information transparency , And build a unified and easy to use visual analysis platform , Improve the efficiency of tabulation . This has become BI New trends in systems .

image.png

be based on MaxCompute Yunshucang +BI Characteristics of

MaxCompute( primary ODPS) It's a big data computing service , It provides flexibility and speed 、 Full custody 、 High performance 、 Low cost 、 Safe PB Level data warehouse solution , It enables you to analyze and process massive data economically and efficiently . be based on MaxCompute The basic architecture of cloud data warehouse is shown in the figure below . The underlying cluster is MaxCompute Built by itself , Users don't need to feel . Up again , There are a variety of computing engines . The engine provides all kinds of API, There is also a deep integration of a one-stop big data intelligent cloud R & D platform DataWorks. Under such a system of cloud data warehouse , Data can be prepared , Do all kinds of cleaning 、 machining 、 After analysis , You can enter a stage of data consumption .

image.png

To sum up MaxCompute The characteristics of cloud data warehouse . First of all , It's an out of the box online service . No platform operation and maintenance , Low total cost of ownership . second , Ultimate flexibility . Elastic expansion , Respond to rapid changes in business size without capacity planning . Third , Simple and easy to use , Multi functional computing services . Various computing models , Multiple data channels , External data source federated computing . Fourth , Enterprise security capabilities . Multi tenant security mechanism , Fine grained authorization , Data encryption 、 desensitization , Backup recovery . The fifth , Ecological integration . Support multiple data sources 、 Ecological tools and standards .

image.png

be based on MaxCompute Cloud data warehouse , We and BI How do tools dock .MaxCompute It's mainly a storage and computing service , Plus a data development platform DataWorks, It forms an offline cloud data warehouse . On top of that , Deep integration of Alibaba cloud's Quick BI. It's an analytical reporting tool , Connect directly to one MaxCompute You can analyze this table by yourself . There are also some third-party tools , Velvety ,Tableau. At the same time, we are on the ecological side ,JDBC It also supports . There are also some enterprises 、 Some customers have more diversified needs or personality needs for business intelligence , The existing docking tools may not support , So it can also be done through SDK The way to connect , So as to achieve the goal based on MaxCompute Cloud data warehouse docking a business intelligence information platform .

image.png

Let's see MaxCompute How to realize a high performance and low latency analysis query in offline data warehouse . It can read the offline data warehouse directly , Support diversified query analysis , It includes some simple queries 、 Complex query 、 Point query 、 Federal queries and so on . It can also have rich data sources at the bottom , adopt MaxCompute + Hologres Form an interactive analysis . In such a big data environment , It can be seamless docking . for instance Quick BI,Tableau, Velvety . So it can get started very quickly , Through such a combination, we can quickly realize an enterprise information platform .

image.png

Practical cases

Next, let's take a look at a few practical cases .

An industry case of new retail , Demand background : be based on Hadoop Open source ecological creation , Hardware and software maintenance costs are high , Stability issues continue , Serious impact on business analysis ; Online business explosion , There is a huge backlog of demand , Expect an overall solution , Be able to quickly and flexibly support the technical expansion required by business development . Through such a big data solution , I directly used Alibaba cloud's Quick BI This product , It has realized the rapid transformation of digital intelligence , Embrace new retail , Reduce TCO At the same time , Better rely on cloud Ecology , Realize the closed loop of data assets business . Finally, the new retail case , Based on our MaxCompute + DataWorks, Improved his data business development efficiency .

image.png

Let's take another example of new finance . Demand background : Financial business data , There are strong requirements for safety control , We need a complete safety management 理 system , At the same time, personalized security needs should be met ; Business is growing fast , Need to be able to build quickly 、 The cost is low 、 Second level expansion of the data platform system . The value we create for our customers : be based on MaxCompute Out of the box applications meet their data security requirements in the security audit process , To shorten the 了 Demand response time and meet its personalized requirements on data security .

image.png

 

Link to the original text
This article is the original content of Alibaba cloud , No reprint without permission .

版权声明
本文为[Aliyun yunqi]所创,转载请带上原文链接,感谢
https://chowdera.com/2020/11/20201112221016759w.html