MaxCompute is widely used in various fields to process big data in the cloud. MaxCompute solves issues that are related to the analysis of large amounts of data and reduces O&M costs of enterprises. This way, enterprises can concentrate on business development. This topic describes featured customer stories of MaxCompute.

Alibaba economy: Umeng

  • Customer profile

    Umeng is an independent third-party global data intelligence service provider that uses technology and algorithm capabilities to mine labels and analyze metrics based on global data resources. Umeng helps enterprises gain deep user insights to drive real-time business decision-making and continuous business growth.

  • Customer demands
    • Help enterprises and developers integrate and analyze data of independent data systems.
    • Help enterprises and developers achieve a balance between the flexibility of business intelligence (BI) systems and business availability.
  • Solution

    Umeng and MaxCompute jointly build a developer databank to provide the self-service analysis service Data Open Platform (U-DOP). U-DOP fully integrates the data of Umeng with the data of enterprises. U-DOP returns data to MaxCompute by subscribing to data packages, presets analysis templates, and analyzes data by using BI tools. This way, U-DOP provides enterprises with flexible end-to-end data analysis capabilities. The following figure shows the architecture of the solution.

    Architecture

E-commerce: Wanwudezhi

  • Customer profile

    Wanwudezhi is an online platform for trading Chinese handicrafts and antiques. This platform provides an online antique appraisal service and an online space for the traditional Chinese culture community. The number of monthly active users on this platform reaches 6,000,000. Wanwudezhi provides three core services: live broadcast, auction, and Buy It Now. Wanwudezhi follows the "appraisal first, delivery later" principle and builds a transparent, healthy, and secure cultural consumption environment for users. This way, consumers can buy traditional Chinese items with high quality and feel the charm of traditional Chinese culture.

  • Customer demands

    Wanwudezhi wants to use the SaaS and PaaS services provided by the cloud platform to build a research and development system that can help improve development efficiency and reduce labor costs. Wanwudezhi also wants to efficiently migrate the original MySQL system to the cloud.

  • Solution
    A big data platform is built for Wanwudezhi based on the Alibaba Cloud DataWorks and MaxCompute frameworks. This platform provides core storage and computing components, and upper-level visualization and business query capabilities for Wanwudezhi. Custom software development is performed based on the original open source solution. The following figure shows the architecture of the solution. Architecture

Internet-based social media: Xiaodaka

  • Customer profile

    Xiaodaka is an interest community program that offers interest circles such as painting, yoga, fitness, photography, parenting, reading, and designer toys. Xiaodaka allows users to share things, communicate, and grow together in the circles that they are interested in. Xiaodaka has millions of daily active users and can generate terabytes of data every day.

  • Customer demands

    Xiaodaka wants to build a data warehouse that features agile development and high extensibility at low development costs and O&M costs.

  • Solution

    A MaxCompute data warehouse is built at low labor costs and software and hardware costs. The data warehouse ensures high resource utilization. The deployment process of the data warehouse is easy and can be complete by only one developer. The following figure shows the architecture of the solution.

    Architecture

Internet finance: Tianhong Asset Management

  • Customer profile

    Tianhong Asset Management is the largest public fund manager in China and aims to become the largest index fund service provider. Tianhong Asset Management innovates its services based on the needs of customers and uses index funds as the underlying tool to provide customers with end-to-end services. Tianhong Asset Management and Alipay jointly launched Yu'E Bao.

  • Customer demands

    As the number of Yu'E Bao users and the amount of data exponentially increase, Tianhong Asset Management can no longer manage data only by using Hadoop clusters. In addition, to make business decisions and predict user behavior, business personnel require accurate user analysis results.

  • Solution

    Tianhong Asset Management builds an enterprise-level end-to-end big data solution based on MaxCompute. MaxCompute has powerful, secure, and stable storage, O&M, and computing capabilities that can be used to store and process large amounts of data. MaxCompute significantly improves the data processing efficiency of Tianhong Asset Management. For example, MaxCompute reduces the data processing time from 8 hours to 1.5 hours. MaxCompute also reduces the workloads of on-premises servers, development costs, and labor costs, which allows developers to focus only on business development. This solution enables Tianhong Asset Management to provide better financial services for users. The following figure shows the architecture of the solution.

    Architecture

Smart delivery: Qianxun Spatial Intelligence

  • Customer profile

    Qianxun Spatial Intelligence (Qianxun SI) is a leading high-precision positioning service provider in the world. Qianxun SI offers centimeter-level dynamic positioning services and millimeter-level static positioning services. Qianxun SI relies on BeiDou Navigation Satellite System and is compatible with the Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), and Galileo. Qianxun SI has deployed more than 2,400 ground-based augmentation stations across China. Qianxun SI provides users across China with high-precision positioning services and extended services by using self-developed positioning algorithms and Internet-based big data computing technologies.

  • Customer demands

    Qianxun SI wants to improve computing precision and speed to meet various requirements for real-time high-precision positioning.

  • Solution

    A solution based on MaxCompute is developed for Qianxun SI to eliminate the need to build self-managed clusters. In the hybrid cloud architecture, confidential data is processed on the Alibaba Cloud hybrid cloud, large-scale data in the cloud is computed in MaxCompute, and positioning data is broadcast on the Alibaba Cloud public cloud.