开源时序数据库 InfluxDB

介绍

InfluxDB™ is a time series database designed to handle high write and query loads. It is an integral component of the TICK stack. InfluxDB is meant to be used as a backing store for any use case involving large amounts of timestamped data, including DevOps monitoring, application metrics, IoT sensor data, and real-time analytics.

基本概念

DataBase

　类似于传统数据库中的 DataBase 概念

Measurement

　和 OLAP 中广义上的度量概念一致，某些 OLAP 数据库中又称为 metric

Tag

　和 OLAP 中广义上的维度概念一致，某些 OLAP 数据库中又称为 tagKV

Field

　数值

Timestamp

　时间戳

Points

　数据点

Series

　数据点组成的序列

Retention Policy

　数据过期策略，即 TTL

环境部署

目前，因为 InfluxDB 2.x 还没有稳定版本发布，所以这里我们使用的 InfluxDB 版本是 1.7.x

编译

$ go get github.com/influxdata/influxdb
$ cd $GOPATH/src/github.com/influxdata/influxdb
$ go clean ./...
$ go install -ldflags="-X main.version=1.7.10" ./...
$ ll $GOPATH/bin
复制

启动

$ $GOPATH/bin/influxd
复制

Grafana

# 安装
$ wget https://dl.grafana.com/oss/release/grafana-6.0.0-beta2.x86_64.rpm
$ sudo yum localinstall grafana-6.0.0-beta2.x86_64.rpm

# 启动
$ systemctl daemon-reload
$ systemctl start grafana-server
$ systemctl status grafana-server

# 自启动
$ sudo systemctl enable grafana-server.service
复制

Telegraf

$ wget https://dl.influxdata.com/telegraf/releases/telegraf-1.12.3-1.x86_64.rpm
$ sudo yum localinstall telegraf-1.12.3-1.x86_64.rpm
$ telegraf config > telegraf.conf
$ telegraf --config telegraf.conf
复制

基本操作

版本

$ curl -sL -I localhost:8086/ping | grep -i version
复制

X-Influxdb-Version: v1.7.10
复制

控制台

# 输入如下命令，即可进入到 InfluxDB 命令交互的控制台
$ influx -host 'localhost' -port '8086'
复制

建表

> show databases;
复制

name: databases
name
----
_internal
复制

# CREATE DATABASE <database_name> [WITH [DURATION <duration>] [REPLICATION <n>] [SHARD DURATION <duration>] [NAME <retention-policy-name>]]
> create database "yuzhouwan";
复制

> show databases;
复制

name: databases
name
----
_internal
yuzhouwan
复制

> use yuzhouwan;
复制

Using database yuzhouwan
复制

写入

> insert blog,protocol=https,name=yuzhouwan value=666
复制

查询

明细查询

> select * from blog
复制

name: blog
time                name      protocol value
----                ----      -------- -----
1556438552229094000 yuzhouwan https    666
复制

聚合查询

> select mean(value) from blog
复制

name: blog
time mean
---- ----
0    666
复制

范围查询

> select * from blog WHERE time > '2019-04-01T00:00:00Z' OR time < '2019-05-01T00:00:00Z'
复制

name: blog
time                name      protocol value
----                ----      -------- -----
1556438552229094000 yuzhouwan https    666
复制

数据导出

query 接口

$ curl -G 'http://localhost:8086/query?pretty=true' --data-urlencode "db=yuzhouwan" --data-urlencode 'q=select * from blog'
复制

{
  "results": [
    {
      "series": [
        {
          "columns": [
            "time",
            "name",
            "protocol",
            "value"
          ],
          "name": "blog",
          "values": [
            [
              "2019-04-28T08:02:32.229094Z",
              "yuzhouwan",
              "https",
              666
            ]
          ]
        }
      ],
      "statement_id": 0
    }
  ]
}
复制

influx_inspect 命令

$ influx_inspect export -database yuzhouwan -start 2019-01-01T00:00:00+00:00 -end 2019-12-01T00:00:00+00:00  -out yuzhouwan.out
复制

writing out tsm file data for yuzhouwan/autogen...complete.
复制

$ cat yuzhouwan.out
复制

# INFLUXDB EXPORT: 2019-01-01T08:00:00+08:00 - 2019-12-01T08:00:00+08:00
# DDL
CREATE DATABASE yuzhouwan WITH NAME autogen
# DML
# CONTEXT-DATABASE:yuzhouwan
# CONTEXT-RETENTION-POLICY:autogen
# writing tsm data
blog,name=yuzhouwan,protocol=https value=666 1556438552229094000
复制