Amazon AWS

 the AWS cloud services adopted by the first thermal power company include Amazon Elastic Compute Cloud (Amazon EC2), Amazon Simple Storage Service (Amazon S3), Amazon relational database service (Amazon RDS), Amazon VPC, AWS Iam, Amazon cloudwatch, AWS cloudtrail , business support, etc.
 
 User: xingyin information technology (Shanghai) Co., Ltd.
 
 xiaohongshu is a new generation of community e-commerce, which combines overseas shopping sharing community with cross-border e-commerce, accurately captures the consumption upgrading needs of post-85 and post-90 generations, and rapidly develops into an influential global shopping sharing community. At present, the number of registered users of xiaohongshu has exceeded 100 million. In the first half year of the establishment of the e-commerce platform, its sales reached 700 million yuan, which was praised by Premier Li Keqiang as "one of the fastest growing start-ups in China". The little red book app also launched the "Black Star" in November 2015During the "phase 5" period, it ranked first in the app store for two consecutive days.

 the key to the success of xiaohongshu mode is that the technical team of xiaohongshu can understand what users like, what they are sharing and what they like most through the background data? And through the analysis of these data, we can speculate which commodities may be popular. Therefore, the little red book needs to deal with data from different data sources: structured data from e-commerce; unstructured data from the community, such as pictures, comments, facial expressions, which are difficult to recognize by machines; and massive log data. 

Therefore, the data team needs to quickly build a data processing system to clean and reorganize the data efficiently to make it suitable for data analysts. Amazon EMR, Amazon RDS, Amazon redshift and other services provided by AWS cloud platform constitute a perfect data processing infrastructure. In particular, Amazon redshift is a fast, fully hosted Pb level data warehouse solution, which makes it very simple and cost-effective to analyze data with existing business intelligence tools. The data team of little red book has made full use of redshift. At first, little red book used Amazon S3 and Amazon redshift to synchronize the transaction data in the business database, and then began to use Amazon S3, Amazon EMR and Amazon redshift to parse and update user logs At the same time, Amazon redshift, as a data warehouse, undertakes a lot of follow-up Bi analysis work.

the efficiency improvement brought by using AWS data processing service is obvious. If there is no AWS, it takes several engineers to build a usable log analysis system, which takes several months to complete. With the data service of AWS, the two engineers of xiaohongshu set up the whole system in two weeks and formally used it in the production environment. In addition, the speed of solving new data requirements is also significantly improved. 

Compared with writing code to pull data from the business back-end database every time, SQL query from the data warehouse significantly improves the efficiency, reducing the workload of the past few hours to a few minutes. Most importantly, the cloud service of AWS makes it a reality for little red book to build a "small and beautiful data team",   so that it can quickly try new models and new data processing methods, and find the most suitable data solution. In addition, the use of AWS has greatly changed the work style of the little red book team. It has become a data-driven team. No matter it is the e-commerce, community, or even marketing team, they all carry out business under the guidance of data. The company's decision-making and performance evaluation also depend on data.
 
 User: Guangzhou Huiliang Network Technology Co., Ltd. (mobvista)  
 
mobvista is the world's leading mobile advertising platform. The group's business includes the global mobile advertising network and overseas distribution game platform. Mobvista adheres to the concept of technology innovation driven and customized service, and focuses on providing the most efficient and high-quality user access and flow cashing solutions for global partners. At present, mobvista's mobile advertising network has covered 236 countries around the world, with a daily display volume of more than 10 billion. It ranks the 12th in the list of global advertisers' partners released by tune, and the third in the comprehensive ranking of global Android game traffic released by appsflyer, an authoritative third-party monitoring platform.

"we migrated to the cloud platform of AWS when the system scale was still very small, which not only enabled us to obtain the required system stability, but also made the system architecture more and more complex with the development of business. AWS can provide solutions matching our architecture for rapid deployment and smooth upgrade. "   Said Huang Weijian, co-founder and technology VP of mobvista.

mobvista has adopted a step-by-step approach in using AWS services. Currently, mobvista has adopted AWS services in nine regions, including Amazon EC2, elastic load balancing, Amazon redshift, Amazon dynamodb, Amazon S3, Amazon RDS, Amazon EMR and a series of products, covering 236 cities around the worldNational and regional mobile advertising network platform. "We migrated to the cloud platform of AWS when the system scale was very small, which not only enabled us to obtain the required  cissp pocket prep  system stability, but also made the system architecture more and more complex with the development of business. AWS can provide solutions matching our architecture, rapid deployment and smooth upgrade. "   Huang Weijian said.

mobvista finally chose Amazon dynamodb, the cloud dynamic database of AWS, which allows users to store and retrieve any amount of data in a simple and cost-effective way, and its reading and writing can be adjusted. Users can use the program to monitor the traffic of Amazon dynamodb, and then configure it automatically on demand, "Amazon dynamodb meets our needs very well, with good results: high speed, strong stability and low maintenance cost. At present, no other cloud service provider can provide similar services." Huang Weijian said.

in terms of offline statistical analysis, AWS redshift's powerful function and flexible billing mode make it the best choice for mobvista. For example, a click data contains 10-20 dimensions of information: user IP, browser type, version of operating system, which advertisement is clicked, what information is included in the advertisement, etc. with 100-200 million hits per day, it is almost impossible to aggregate analysis by traditional methods. R has AWSEdshift, no matter which dimensions need to be analyzed, can be completed very conveniently and quickly. "AWS's unique technology products bring us changes from 'impossible' to 'high performance'" Huang Weijian said.