It is very common for big data capabilities to be provided in the form of cloud services, such as aws, azure, domestic Alibaba, and Huawei all have similar services. Today we discuss several models of big data cloud services.
The first, the most typical is called cluster hosting mode
The most typical service is the emr service of aws, which can be accessed at: https://aws.amazon.com/cn/elasticmapreduce/. In this mode, the cloud service mainly solves the installation, monitoring, operation and maintenance management of big data component clusters, etc., which reduces the technical knowledge threshold of technical personnel to the bottom layer of big data clusters. Cloud service resources are applied for on-demand, which speeds up the time for business deployment, and at the same time converts one-time procurement costs into on-demand costs, reducing the risk of enterprise operations.
Under this mode, the core is to reduce the burden of operation and maintenance. However, business personnel also need to select different components according to different business requirements, and combine the use of various components. Each component is similar to the accessories in the computer city. Can there be a more thorough way to provide a brand machine to customers?
The second, server-less mode
The term server-less comes from AWS's lambda service (https://aws.amazon.com/cn/lambda/), through AWS Lambda, you can run code without configuring or managing servers. You only pay for the compute time you consume - no charges are incurred when the code is not running. With Lambda, you can run code for almost any type of application or backend service, all without administration. Just upload your code, and Lambda takes care of everything needed to run and scale your highly available code. You can set up your code to automatically trigger from other AWS services, or call directly from any web or mobile application.
The core of this is that customers don't need to care about the server, they only need to care about their own code. The field of big data can also provide services with similar concepts.
这种模式最典型的是azure的data-lake。可以访问https://azure.microsoft.com/zh-cn/solutions/data-lake/。客户不用再关心各个独立的组件。Azure Data Lake 包括了所有所需的功能，使开发人员、数据专家和分析师可以更轻松地存储任何大小、形状和速度的数据以及跨平台和语言进行各种类型的处理和分析。它消除了插入和存储所有数据的复杂性，同时启动更快，可与批量、流式、交互式分析一起运行。
这种服务底层是构建在基础的托管集群上，把各种服务组合在一起，提供统一的访问。服务本身提供自动弹性伸缩的能力，数据自动搬迁，自动保障客户的SLA。
第三种，大数据saas服务
品牌机相对电脑散件肯定更易用，但是对完全没有电脑知识的人也很难用，而去很多用电脑的人核心是要使用word之类的office软件，更不关心电脑本身其他更多的功能。所以大数据最重要能根据行业提供saas类的服务或者应用。阿里云最近推出了一个郡县图治，有点意思：
总结一下
讲到最后，客户的层次，和处于业务的情况是不一样的，所以需求也是多种的，因此这三种服务模式都有广阔的空间。第一种模式想对而已，已经陷入同质化竞争，创新的空间有限，未来更看好第二种，第三种模式的发展。
微信扫一扫
关注该公众号