Skip to main content

TDEngine

Chaitanya Diwadkar​

  1. Github

    There are some novel concepts in TDengine which give us the performance and compression gains that we have over TimeScale and Influx.

  2. Cloud - let me know if you have any issues signing up or anything. No credit card is required and in the free tier we keep your data for 30 days. During instance creation it gives you the option of creating a sample database with synthetic data from electrical meters.

Installing TDEngine Service​

https://docs.tdengine.com/get-started/package/

OS: Ubuntu 20.04

wget -qO - http://repos.taosdata.com/tdengine.key | sudo apt-key add -
echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-stable stable main" | sudo tee /etc/apt/sources.list.d/tdengine-stable.list

sudo apt-get update
apt-cache policy tdengine
sudo apt-get install tdengine

On the first node in your TDengine cluster, leave the Enter FQDN: prompt blank and press Enter. On subsequent nodes, you can enter the endpoint of the first dnode in the cluster. You can also configure this setting after you have finished installing TDengine.

Running the service​

  1. To start the service systemctl start taosd
  2. Check the status systemctl status taosd

This can be interfaced with TCEngine CLI, REST API or language client drivers

TDEngine C API​

C Connector​

Testing : https://github.com/gqc/TDEngine_trial

Example for RESTful API: tqdemo.c What is curl? link

Information on the database used for Sensor Mesh project​

Database specifications​

We have to decide the following

  • Number of days to retain data (eg: 365 days)
  • File partitioning interval (eg: Make a new file every 10 days)
  • Write buffer size (Default is 96MB. Have to adjust based on the performance
  • Check the all available options.

Creating the database​

Using taos CLI

  1. Open the CLI

    taos

  2. Example Setup the database using SQL commands (details)

    database design

    -- Create Database
    CREATE DATABASE test; -- There are optional arguments: KEEP 365 DURATION 10 BUFFER 16 WAL_LEVEL 1;

    -- Set the current table to the table we just created
    USE test;

    -- Create supertable
    CREATE STABLE sensor_nodes_timeseries (ts timestamp, acceleration_gx binary(4096), acceleration_gy binary(4096), acceleration_gz binary(4096)) TAGS (MAC_ID binary(20));

    -- Create table
    -- The table name 'sensor_1' has to be replaced with a unique_id, probably MAC-ID
    -- TAGS argument has to be the MAC-ID of the node as per our design
    -- Every node in the system has to have a table of it's own
    CREATE TABLE sensor_1 USING sensor_nodes_timeseries TAGS (201545654);

    -- Alternatively the table can be created on the fly with the first data entry
    -- If the table is already present, this will just append the new data into it
    INSERT INTO sensor_2 USING sensor_nodes_timeseries TAGS (201545654) VALUES (now,02154564, 2224, 2555);

Tables used for displaying timeseries and FFT with grafana​

CREATE TABLE IF NOT EXISTS current_timeseries_data (ts timestamp, time_dt int, amplitude_gx int, amplitude_gy int, amplitude_gz int)
CREATE TABLE IF NOT EXISTS current_serverFFT_data (ts timestamp, f_bin float, fft_gx float, fft_gy float, fft_gz float)

Backing up TDEngine data​

  1. taosDump
    1. Install following instructions here
    2. taosdump -D <DB_NAME>