Machine Learning Flow
Architecture for Machine Learning Basic Flow
About Architecture
Artificial intelligence (AI) has been developed for finding patterns by itself and making predictions on the future based on big data along with the development of computer technology. To apply artificial intelligence (AI) to business, collection and analysis on the data of the past must precede it. You can apply the machine learning algorithm on the user experience based data or the industry-characterized mass data to derive more specific data, predict future behavior, or derive reliable results on the speed and quality of logistics, and make appropriate decisions based on these. The cloud environment is very effective because you do not have to configure the new infrastructure for machine learning yourself and can instantly configure and utilize the basic infrastructure for machine learning. You can use the service at a reasonable price considering that you will be able to use it as needed and make payments on the amount you have used using the managed products for big data analysis or the storage products to store/manage the mass data.
Architecture
9-1_MachineLearningFlow_1601269085396.png)
Related Services
Use Cases and Effect
- Create Clusters Easily and Simply
- Cloud Hadoop automatically creates clusters to ease the burden of resources required for infrastructure management tasks. You can have a system that can be analyzed at any time throughout the installation, configuration, and optimization of several open-source frameworks. Open-source frameworks, such as Hadoop, HBase, Spark, and Hive, are installed, and clusters with optimized configurations are created, allowing users to perform the tasks needed for analysis right away.
- Securing Flexible Scalability and High Availability
- To ensure high availability, two master nodes are provided for redundancy when creating a Cloud Hadoop cluster. The role of standby node changes if the master node fails so that the performance of the role as master node is possible, and the number of instances required for data analysis can be easily decreased or increased from 1 to 8 at a time desired by the user.
- Web UI for Cluster Management and Monitoring
- Enjoy a convenient UI for managing information and the status for Cloud Hadoop clusters. Apache Ambari, an open-source application, makes managing and monitoring Cloud Hadoop clusters easy and efficient by leveraging the simplicity of the web UI and REST API. Also, you can freely configure Hadoop, HBase, Spark and Hive, etc. even without logging in to the server directly.
- Unlimited Object Storage Based Data Capacity
- Save large amount of data at a low cost using Object Storage on the NAVER CLOUD PLATFORM as a data storage service. You can use without worrying about the capacity since easy extension is possible at a reasonable cost from a gigabyte unit to a petabyte unit according to the customer's business scale, and you can also link to analyze data in Cloud Hadoop.
- Choose a Computer Power of Your Need
- Since the servers with various types of computing power are provided, the user is able to analyze mass data quickly by selecting various servers according to the performance required for the analysis.