Axibase Time Series Database

SQL Report Metadata

Overview

The metadata section is located in the header and is prepended with the hash symbol #. It includes the following fields in the "name,value" format:

Name Description
publisher Name of the reporting product (ATSD) and the URL of the ATSD server.
created Date when the report is produced in ISO format.
title Report name, such as "CPU Busy Daily Report".
command Query statement on one or multiple lines.

In addition, the metadata header contains a list of column names with their respective data types.

#name,entity,Average,...
#datatype,string,double,...

Metadata Specification

Metadata in CSV Format

Metadata can be embedded in CSV files as part of the header using the hash symbol (#).

#publisher,Axibase Time Series Database,https://atsd.example.org
#created,2016-06-12T15:56:39.106Z
#title,SQL Query
#comment,"SELECT entity, avg(value) AS "Average", median(value), max(value), count(*)
#FROM "mpstat.cpu_busy"
#  WHERE time BETWEEN previous_day and current_day
#  GROUP BY entity
#  ORDER BY avg(value) DESC"
#name,entity,Average,median(value),max(value),count(*)
#datatype,string,double,double,double,double
entity,Average,median(value),max(value),count(*)
nurswgvml006,18.4,4.0,100.0,5385.0

Metadata in Excel Format

Same as in CSV format.

Metadata in JSON Format

Metadata is specified according to W3C Recommendation Metadata Vocabulary for Tabular Data

Table schema object provides the following information about the columns in the result set:

  • index, starts with 1
  • name
  • label (alias)
  • datatype
  • table name (metric name)
  • type as specified in ontology
  • description
{
  "metadata": {
    "@context": ["http://www.w3.org/ns/csvw", {
      "atsd": "http://www.axibase.com/schemas/2017/07/atsd.jsonld"
    }],
    "dc:created": {
      "@value": "2017-07-04T17:08:39.745Z",
      "@type": "xsd:date"
    },
    "dc:publisher": {
      "schema:name": "Axibase Time Series Database",
      "schema:url": {
        "@id": "https://atsd.example.org"
      }
    },
    "dc:title": "SQL Query",
    "rdfs:comment": "SELECT entity, datetime, avg(value) AS \"Average\" FROM \"mpstat.cpu_busy\" WHERE datetime > current_minute GROUP BY entity, period(1 minute) ORDER BY avg(value) DESC",
    "@type": "Table",
    "url": "sql.csv",
    "tableSchema": {
      "columns": [{
        "columnIndex": 1,
        "name": "entity",
        "titles": "entity",
        "datatype": "string",
        "table": "mpstat.cpu_busy",
        "propertyUrl": "atsd:entity"
      }, {
        "columnIndex": 2,
        "name": "datetime",
        "titles": "datetime",
        "datatype": "xsd:dateTimeStamp",
        "table": "mpstat.cpu_busy",
        "propertyUrl": "atsd:datetime",
        "dc:description": "Sample time in ISO8601 format"
      }, {
        "columnIndex": 3,
        "name": "avg(value)",
        "titles": "Average",
        "datatype": "double",
        "table": "mpstat.cpu_busy",
        "propertyUrl": "atsd:avg"
      }]
    }
  },
  "data": [
    ["nurswgvml007", "2017-07-04T17:08:00.000Z", 11.756666666666664],
    ["nurswgvml006", "2017-07-04T17:08:00.000Z", 3.3499999999999996],
    ["nurswgvml502", "2017-07-04T17:08:00.000Z", 2.9966666666666666],
    ["nurswgvml010", "2017-07-04T17:08:00.000Z", 0.375],
    ["nurswgvml301", "2017-07-04T17:08:00.000Z", 0.0]
  ]
}