Types of Instance

Last updated: 2019-07-25 17:31:13

PDF

To meet the needs of different customers in various application scenarios, Tencent Cloud provides the following recommendations for selecting an instance type:

  • Personal Website

    Standard instances are recommended for general workloads, such as medium- and small-sized Web applications and databases.

  • Enterprise Website/E-commerce/App

    Standard instances are recommended for general workloads, such as medium- and small-sized Web applications and databases.

  • Relational Database/Distributed Cache

    MEM Optimized instances are recommended for scenarios that require extensive memory operations, searches, and computing.

  • NoSQL Database

    High IO instances are recommended for I/O-intensive scenarios that require high disk read/write performance and low latency, such as NoSQL databases (e.g. MongoDB) and clustered databases.

  • High Performance Compute

    Computing or Computing Network Enhanced instances are recommended for scenarios that require a large number of computing resources, such as large client games, high performance science and engineering applications, and video encoding/decoding.

  • High-Performance Client Games

    Computing or Computing Network Enhanced instances are recommended for scenarios that require a large number of computing resources, such as large client games, high performance science and engineering applications, and video encoding/decoding.

  • Mobile/Browser Games

    Computing or Computing Network Enhanced instances are recommended for scenarios that require a large number of computing resources, such as large client games, high performance science and engineering applications, and video encoding/decoding.

  • LVB

    Standard Network Enhanced or Computing Network Enhanced instances are recommended, which come with a 25 GB ENI that is 2.5 times faster than that of regular ten-gigabit data centers, providing a larger bandwidth and a lower latency.

  • Finance

    CDH Standard instances are recommended. Compared with standard instances, these instances provide exclusive physical servers, which ensures the isolation of resources. They are secure, controllable, and in full compliance with the strict regulations in the finance industry. Custom specifications are also supported.

  • Scientific Computing

    GPU Computing instances are recommended for deep learning, and scientific computing including computational fluid dynamics, computational finance, genomics research, environmental analysis, high-performance computing, and other server-side GPU computing workloads.

  • Machine Learning

    GPU Computing instances are recommended for deep learning, and scientific computing including computational fluid dynamics, computational finance, genomics research, environmental analysis, high-performance computing, and other server-side GPU computing workloads.

  • Rendering

    GPU Rendering instances are recommended for non-linear editing, video encoding/decoding, graphics acceleration visualization, and 3D design.

  • Hadoop/Spark/Elastic Search

    Big Data instances are recommended for distributed computing services like Hadoop (HDFS/MapReduce/Spark/Hive), massive parallel processing (MPP) data warehouses, B8 logs, data processing applications.