Skip to main content

Get Started with TDengine TSDB Using an Installation Package

You can install TDengine TSDB on Linux and Windows. To install TDengine TSDB in a Docker container instead of on your machine, see Get Started with TDengine TSDB in Docker.

Before You Begin​

  • Verify that your machine meets the minimum system requirements for TDengine TSDB. For more information, see Supported Platforms and System Requirements.
  • (Windows only) The installer checks for Microsoft Visual C++ Redistributable 2015-2022 x64 version 14.44 or later. If it is missing or outdated, the installer can install the bundled 14.50.35719.0 redistributable after you confirm the prompt. If the installation fails, download the latest supported package from Microsoft Visual C++ Redistributable latest supported downloads.

Procedure​

  1. Download the tar.gz installation package from the list below:

    • Navigate to the directory where the package is located and extract it using tar. For example, on an x64 architecture:

      tar -zxvf tdengine-tsdb-enterprise-{{VERSION}}-linux-x64.tar.gz

    • After extracting the files, go into the subdirectory and run the install.sh script:

      sudo ./install.sh

    For more package types and versions, visit the TDengine Download Center.

    Start the Service​

    After installation, execute the following script in your terminal to start all services:

    sudo start-all.sh

    All TDengine TSDB components are managed by systemd. You can check their service status with the following commands:

    sudo systemctl status taosd
    sudo systemctl status taosadapter
    sudo systemctl status taoskeeper
    sudo systemctl status taos-explorer

    If the output shows the status as Active: active (running) since ..., it means the services have started successfully.

    Quick Start​

    Open the Command Line​

    TDengine provides the taos command-line tool for checking service status, running SQL, and managing databases and tables. After you enter the shell, you can try write and query operations immediately.

    On Linux or macOS, run:

    taos

    On Windows, run:

    taos.exe

    After you enter the shell, the prompt looks like:

    taos>

    In the shell, each SQL statement must end with a semicolon (;). The following example creates a database, inserts two rows, and queries the result:

    CREATE DATABASE demo;
    USE demo;
    CREATE TABLE t (ts TIMESTAMP, speed INT);
    INSERT INTO t VALUES ('2019-07-15 00:00:00', 10);
    INSERT INTO t VALUES ('2019-07-15 01:00:00', 20);
    SELECT * FROM t;

               ts            |    speed    |
    ========================================
     2019-07-15 00:00:00.000 |          10 |
     2019-07-15 01:00:00.000 |          20 |

    Query OK, 2 row(s) in set (0.003128s)

    Besides SQL, you can use the shell to check system status and manage users. The CLI and client drivers can also be installed separately on other machines. For details, see TDengine CLI.

    Write Test Data​

    taosBenchmark is a TDengine testing tool for generating sample data and trying write, query, and subscription features. It can simulate data from many devices and supports configuring the database, supertable, tag columns, data columns, subtable count, records per subtable, write interval, worker threads, and whether to write out-of-order data.

    After confirming that the TDengine service is running, run the following command in a terminal:

    taosBenchmark -y

    This automatically creates a supertable named meters in the test database. The supertable contains 10,000 subtables named d0 through d9999, each with 10,000 records. Each record has four fields: ts (timestamp), current, voltage, and phase. Timestamps range from 2017-07-14 10:40:00.000 to 2017-07-14 10:40:09.999. Each table also has two tags, location and groupId, where groupId is from 1 to 10 and location includes California cities such as California.Campbell and California.Cupertino.

    The command writes 100 million records. Elapsed time depends on hardware; even on a typical PC server, it usually takes only about ten seconds.

    taosBenchmark provides many options for customizing table count, record count, and other test settings. Run the following command to list all parameters:

    taosBenchmark --help

    For detailed usage, see taosBenchmark Reference.

    Query Test Data​

    After writing data with taosBenchmark, run the following queries in the shell to experience query performance.

    1. Count all records under the meters supertable:
    SELECT COUNT(*) FROM test.meters;

         count(*)     |
    ===================
            100000000 |

    Query OK, 1 row(s) in set (0.075247s)
    1. Compute the average, maximum, and minimum over 100 million records:
    SELECT AVG(current), MAX(voltage), MIN(phase) FROM test.meters;

       avg(current)   | max(voltage) | min(phase) |
    ===============================================
     10.2084751316071 |          258 |        145 |

    Query OK, 1 row(s) in set (0.094037s)
    1. Count records where location = "California.SanFrancisco":
    SELECT COUNT(*) FROM test.meters WHERE location = "California.SanFrancisco";

         count(*)     |
    ===================
             10340000 |

    Query OK, 1 row(s) in set (0.050540s)
    1. Compute the average, maximum, and minimum for all records where groupId = 10:
    SELECT AVG(current), MAX(voltage), MIN(phase) FROM test.meters WHERE groupId = 10;

       avg(current)   | max(voltage) | min(phase) |
    ===============================================
     10.2084751316071 |          258 |        145 |

    Query OK, 1 row(s) in set (0.033754s)
    1. Aggregate table d1001 every 10 seconds for average, maximum, and minimum:
    SELECT _wstart, AVG(current), MAX(voltage), MIN(phase)
    FROM test.d1001 INTERVAL(10s);

             _wstart         |   avg(current)   | max(voltage) | min(phase) |
    =========================================================================
     2017-07-14 10:40:00.000 | 10.2084751316071 |          258 |        145 |

    Query OK, 1 row(s) in set (0.004716s)

    In the query above, the system pseudocolumn _wstart is the start time of each window.

    Go Further​

    After finishing the steps above, continue with:

    What to Do Next​

    After this quick start, continue with later chapters to learn more about TDengine data modeling, writing, querying, data subscription, and stream processing.