druid/ingestion-spec.md at 6e2eded277bec907466ce0b7823ce70fae6e388e

27 KiB

Raw Blame History

id	title	sidebar_label	description
ingestion-spec	Ingestion spec reference	Ingestion spec	Reference for the configuration options in the ingestion spec.

All ingestion methods use ingestion tasks to load data into Druid. Streaming ingestion uses ongoing supervisors that run and supervise a set of tasks over time. Native batch and Hadoop-based ingestion use a one-time task. All types of ingestion use an ingestion spec to configure ingestion.

Ingestion specs consists of three main components:

dataSchema, which configures the datasource name, primary timestamp, dimensions, metrics, and transforms and filters (if needed).
ioConfig, which tells Druid how to connect to the source system and how to parse data. For more information, see the documentation for each ingestion method.
tuningConfig, which controls various tuning parameters specific to each ingestion method.

Example ingestion spec for task type index_parallel (native batch):

{
  "type": "index_parallel",
  "spec": {
    "dataSchema": {
      "dataSource": "wikipedia",
      "timestampSpec": {
        "column": "timestamp",
        "format": "auto"
      },
      "dimensionsSpec": {
        "dimensions": [
          "page",
          "language",
          { "type": "long", "name": "userId" }
        ]
      },
      "metricsSpec": [
        { "type": "count", "name": "count" },
        { "type": "doubleSum", "name": "bytes_added_sum", "fieldName": "bytes_added" },
        { "type": "doubleSum", "name": "bytes_deleted_sum", "fieldName": "bytes_deleted" }
      ],
      "granularitySpec": {
        "segmentGranularity": "day",
        "queryGranularity": "none",
        "intervals": [
          "2013-08-31/2013-09-01"
        ]
      }
    },
    "ioConfig": {
      "type": "index_parallel",
      "inputSource": {
        "type": "local",
        "baseDir": "examples/indexing/",
        "filter": "wikipedia_data.json"
      },
      "inputFormat": {
        "type": "json",
        "flattenSpec": {
          "useFieldDiscovery": true,
          "fields": [
            { "type": "path", "name": "userId", "expr": "$.user.id" }
          ]
        }
      }
    },
    "tuningConfig": {
      "type": "index_parallel"
    }
  }
}

The specific options supported by these sections will depend on the ingestion method you have chosen. For more examples, refer to the documentation for each ingestion method.

You can also load data visually, without the need to write an ingestion spec, using the "Load data" functionality available in Druid's web console. Druid's visual data loader supports Kafka, Kinesis, and native batch mode.

`dataSchema`

The dataSchema spec has been changed in 0.17.0. The new spec is supported by all ingestion methods except for Hadoop ingestion. See the Legacy dataSchema spec for the old spec.

The dataSchema is a holder for the following components:

datasource name
primary timestamp
dimensions
metrics
transforms and filters (if needed).

An example dataSchema is:

"dataSchema": {
  "dataSource": "wikipedia",
  "timestampSpec": {
    "column": "timestamp",
    "format": "auto"
  },
  "dimensionsSpec": {
    "dimensions": [
      "page",
      "language",
      { "type": "long", "name": "userId" }
    ]
  },
  "metricsSpec": [
    { "type": "count", "name": "count" },
    { "type": "doubleSum", "name": "bytes_added_sum", "fieldName": "bytes_added" },
    { "type": "doubleSum", "name": "bytes_deleted_sum", "fieldName": "bytes_deleted" }
  ],
  "granularitySpec": {
    "segmentGranularity": "day",
    "queryGranularity": "none",
    "intervals": [
      "2013-08-31/2013-09-01"
    ]
  }
}

`dataSource`

The dataSource is located in dataSchema → dataSource and is simply the name of the datasource that data will be written to. An example dataSource is:

"dataSource": "my-first-datasource"

`timestampSpec`

The timestampSpec is located in dataSchema → timestampSpec and is responsible for configuring the primary timestamp. An example timestampSpec is:

"timestampSpec": {
  "column": "timestamp",
  "format": "auto"
}

Conceptually, after input data records are read, Druid applies ingestion spec components in a particular order: first flattenSpec (if any), then timestampSpec, then transformSpec, and finally dimensionsSpec and metricsSpec. Keep this in mind when writing your ingestion spec.

A timestampSpec can have the following components:

Field	Description	Default
column	Input row field to read the primary timestamp from. Regardless of the name of this input field, the primary timestamp will always be stored as a column named `__time` in your Druid datasource.	timestamp
format	Timestamp format. Options are: `iso`: ISO8601 with 'T' separator, like "2000-01-01T01:02:03.456" `posix`: seconds since epoch `millis`: milliseconds since epoch `micro`: microseconds since epoch `nano`: nanoseconds since epoch `auto`: automatically detects ISO (either 'T' or space separator) or millis format any Joda DateTimeFormat string	auto
missingValue	Timestamp to use for input records that have a null or missing timestamp `column`. Should be in ISO8601 format, like `"2000-01-01T01:02:03.456"`, even if you have specified something else for `format`. Since Druid requires a primary timestamp, this setting can be useful for ingesting datasets that do not have any per-record timestamps at all.	none

`dimensionsSpec`

The dimensionsSpec is located in dataSchema → dimensionsSpec and is responsible for configuring dimensions. An example dimensionsSpec is:

"dimensionsSpec" : {
  "dimensions": [
    "page",
    "language",
    { "type": "long", "name": "userId" }
  ],
  "dimensionExclusions" : [],
  "spatialDimensions" : []
}

Conceptually, after input data records are read, Druid applies ingestion spec components in a particular order: first flattenSpec (if any), then timestampSpec, then transformSpec, and finally dimensionsSpec and metricsSpec. Keep this in mind when writing your ingestion spec.

A dimensionsSpec can have the following components:

Field	Description	Default
dimensions	A list of dimension names or objects. Cannot have the same column in both `dimensions` and `dimensionExclusions`. If this and `spatialDimensions` are both null or empty arrays, Druid will treat all non-timestamp, non-metric columns that do not appear in `dimensionExclusions` as String-typed dimension columns. See inclusions and exclusions below for details.	`[]`
dimensionExclusions	The names of dimensions to exclude from ingestion. Only names are supported here, not objects. This list is only used if the `dimensions` and `spatialDimensions` lists are both null or empty arrays; otherwise it is ignored. See inclusions and exclusions below for details.	`[]`
spatialDimensions	An array of spatial dimensions.	`[]`

Dimension objects

Each dimension in the dimensions list can either be a name or an object. Providing a name is equivalent to providing a string type dimension object with the given name, e.g. "page" is equivalent to {"name": "page", "type": "string"}.

Dimension objects can have the following components:

Field	Description	Default
type	Either `string`, `long`, `float`, or `double`.	`string`
name	The name of the dimension. This will be used as the field name to read from input records, as well as the column name stored in generated segments. Note that you can use a `transformSpec` if you want to rename columns during ingestion time.	none (required)
createBitmapIndex	For `string` typed dimensions, whether or not bitmap indexes should be created for the column in generated segments. Creating a bitmap index requires more storage, but speeds up certain kinds of filtering (especially equality and prefix filtering). Only supported for `string` typed dimensions.	`true`
multiValueHandling	Specify the type of handling for multi-value fields. Possible values are `sorted_array`, `sorted_set`, and `array`. `sorted_array` and `sorted_set` order the array upon ingestion. `sorted_set` removes duplicates. `array` ingests data as-is	`sorted_array`

Inclusions and exclusions

Druid will interpret a dimensionsSpec in two possible ways: normal or schemaless.

Normal interpretation occurs when either dimensions or spatialDimensions is non-empty. In this case, the combination of the two lists will be taken as the set of dimensions to be ingested, and the list of dimensionExclusions will be ignored.

Schemaless interpretation occurs when both dimensions and spatialDimensions are empty or null. In this case, the set of dimensions is determined in the following way:

First, start from the set of all root-level fields from the input record, as determined by the inputFormat. "Root-level" includes all fields at the top level of a data structure, but does not included fields nested within maps or lists. To extract these, you must use a flattenSpec. All fields of non-nested data formats, such as CSV and delimited text, are considered root-level.
If a flattenSpec is being used, the set of root-level fields includes any fields generated by the flattenSpec. The useFieldDiscovery parameter determines whether the original root-level fields will be retained or discarded.
Any field listed in dimensionExclusions is excluded.
The field listed as column in the timestampSpec is excluded.
Any field used as an input to an aggregator from the metricsSpec is excluded.
Any field with the same name as an aggregator from the metricsSpec is excluded.
All other fields are ingested as string typed dimensions with the default settings.

Note: Fields generated by a transformSpec are not currently considered candidates for schemaless dimension interpretation.

`metricsSpec`

The metricsSpec is located in dataSchema → metricsSpec and is a list of aggregators to apply at ingestion time. This is most useful when rollup is enabled, since it's how you configure ingestion-time aggregation.

An example metricsSpec is:

"metricsSpec": [
  { "type": "count", "name": "count" },
  { "type": "doubleSum", "name": "bytes_added_sum", "fieldName": "bytes_added" },
  { "type": "doubleSum", "name": "bytes_deleted_sum", "fieldName": "bytes_deleted" }
]

Generally, when rollup is disabled, you should have an empty metricsSpec (because without rollup, Druid does not do any ingestion-time aggregation, so there is little reason to include an ingestion-time aggregator). However, in some cases, it can still make sense to define metrics: for example, if you want to create a complex column as a way of pre-computing part of an approximate aggregation, this can only be done by defining a metric in a metricsSpec.

`granularitySpec`

The granularitySpec is located in dataSchema → granularitySpec and is responsible for configuring the following operations:

Partitioning a datasource into time chunks (via segmentGranularity).
Truncating the timestamp, if desired (via queryGranularity).
Specifying which time chunks of segments should be created, for batch ingestion (via intervals).
Specifying whether ingestion-time rollup should be used or not (via rollup).

Other than rollup, these operations are all based on the primary timestamp.

An example granularitySpec is:

"granularitySpec": {
  "segmentGranularity": "day",
  "queryGranularity": "none",
  "intervals": [
    "2013-08-31/2013-09-01"
  ],
  "rollup": true
}

A granularitySpec can have the following components:

Field	Description	Default
type	Either `uniform` or `arbitrary`. In most cases you want to use `uniform`.	`uniform`
segmentGranularity	Time chunking granularity for this datasource. Multiple segments can be created per time chunk. For example, when set to `day`, the events of the same day fall into the same time chunk which can be optionally further partitioned into multiple segments based on other configurations and input size. Any granularity can be provided here. Note that all segments in the same time chunk should have the same segment granularity. Ignored if `type` is set to `arbitrary`.	`day`
queryGranularity	The resolution of timestamp storage within each segment. This must be equal to, or finer, than `segmentGranularity`. This will be the finest granularity that you can query at and still receive sensible results, but note that you can still query at anything coarser than this granularity. E.g., a value of `minute` will mean that records will be stored at minutely granularity, and can be sensibly queried at any multiple of minutes (including minutely, 5-minutely, hourly, etc). Any granularity can be provided here. Use `none` to store timestamps as-is, without any truncation. Note that `rollup` will be applied if it is set even when the `queryGranularity` is set to `none`.	`none`
rollup	Whether to use ingestion-time rollup or not. Note that rollup is still effective even when `queryGranularity` is set to `none`. Your data will be rolled up if they have the exactly same timestamp.	`true`
intervals	A list of intervals defining time chunks for segments. Specify interval values using ISO8601 format. For example, `["2021-12-06T21:27:10+00:00/2021-12-07T00:00:00+00:00"]`. If you omit the time, the time defaults to "00:00:00". If `type` is set to `uniform`, Druid breaks the list up and rounds-off the list values based on the `segmentGranularity`. If `type` is set to `arbitrary`, Druid uses the list as-is. If `null` or not provided, batch ingestion tasks generally determine which time chunks to output based on the timestamps found in the input data. If specified, batch ingestion tasks may be able to skip a determining-partitions phase, which can result in faster ingestion. Batch ingestion tasks may also be able to request all their locks up-front instead of one by one. Batch ingestion tasks throw away any records with timestamps outside of the specified intervals. Ignored for any form of streaming ingestion.	`null`

`transformSpec`

The transformSpec is located in dataSchema → transformSpec and is responsible for transforming and filtering records during ingestion time. It is optional. An example transformSpec is:

"transformSpec": {
  "transforms": [
    { "type": "expression", "name": "countryUpper", "expression": "upper(country)" }
  ],
  "filter": {
    "type": "selector",
    "dimension": "country",
    "value": "San Serriffe"
  }
}

Conceptually, after input data records are read, Druid applies ingestion spec components in a particular order: first flattenSpec (if any), then timestampSpec, then transformSpec, and finally dimensionsSpec and metricsSpec. Keep this in mind when writing your ingestion spec.

Transforms

The transforms list allows you to specify a set of expressions to evaluate on top of input data. Each transform has a "name" which can be referred to by your dimensionsSpec, metricsSpec, etc.

If a transform has the same name as a field in an input row, then it will shadow the original field. Transforms that shadow fields may still refer to the fields they shadow. This can be used to transform a field "in-place".

Transforms do have some limitations. They can only refer to fields present in the actual input rows; in particular, they cannot refer to other transforms. And they cannot remove fields, only add them. However, they can shadow a field with another field containing all nulls, which will act similarly to removing the field.

Transforms can refer to the timestamp of an input row by referring to __time as part of the expression. They can also replace the timestamp if you set their "name" to __time. In both cases, __time should be treated as a millisecond timestamp (number of milliseconds since Jan 1, 1970 at midnight UTC). Transforms are applied after the timestampSpec.

Druid currently includes one kind of built-in transform, the expression transform. It has the following syntax:

{
  "type": "expression",
  "name": "<output name>",
  "expression": "<expr>"
}

The expression is a Druid query expression.

Conceptually, after input data records are read, Druid applies ingestion spec components in a particular order: first flattenSpec (if any), then timestampSpec, then transformSpec, and finally dimensionsSpec and metricsSpec. Keep this in mind when writing your ingestion spec.

Filter

The filter conditionally filters input rows during ingestion. Only rows that pass the filter will be ingested. Any of Druid's standard query filters can be used. Note that within a transformSpec, the transforms are applied before the filter, so the filter can refer to a transform.

Legacy `dataSchema` spec

The dataSchema spec has been changed in 0.17.0. The new spec is supported by all ingestion methods except for Hadoop ingestion. See dataSchema for the new spec.

The legacy dataSchema spec has below two more components in addition to the ones listed in the dataSchema section above.

input row parser, flattening of nested data (if needed)

`parser` (Deprecated)

In legacy dataSchema, the parser is located in the dataSchema → parser and is responsible for configuring a wide variety of items related to parsing input records. The parser is deprecated and it is highly recommended to use inputFormat instead. For details about inputFormat and supported parser types, see the "Data formats" page.

For details about major components of the parseSpec, refer to their subsections:

timestampSpec, responsible for configuring the primary timestamp.
dimensionsSpec, responsible for configuring dimensions.
flattenSpec, responsible for flattening nested data formats.

An example parser is:

"parser": {
  "type": "string",
  "parseSpec": {
    "format": "json",
    "flattenSpec": {
      "useFieldDiscovery": true,
      "fields": [
        { "type": "path", "name": "userId", "expr": "$.user.id" }
      ]
    },
    "timestampSpec": {
      "column": "timestamp",
      "format": "auto"
    },
    "dimensionsSpec": {
      "dimensions": [
        "page",
        "language",
        { "type": "long", "name": "userId" }
      ]
    }
  }
}

`flattenSpec`

In the legacy dataSchema, the flattenSpec is located in dataSchema → parser → parseSpec → flattenSpec and is responsible for bridging the gap between potentially nested input data (such as JSON, Avro, etc) and Druid's flat data model. See Flatten spec for more details.

`ioConfig`

The ioConfig influences how data is read from a source system, such as Apache Kafka, Amazon S3, a mounted filesystem, or any other supported source system. The inputFormat property applies to all ingestion method except for Hadoop ingestion. The Hadoop ingestion still uses the parser in the legacy dataSchema. The rest of ioConfig is specific to each individual ingestion method. An example ioConfig to read JSON data is:

"ioConfig": {
    "type": "<ingestion-method-specific type code>",
    "inputFormat": {
      "type": "json"
    },
    ...
}

For more details, see the documentation provided by each ingestion method.

`tuningConfig`

Tuning properties are specified in a tuningConfig, which goes at the top level of an ingestion spec. Some properties apply to all ingestion methods, but most are specific to each individual ingestion method. An example tuningConfig that sets all of the shared, common properties to their defaults is:

"tuningConfig": {
  "type": "<ingestion-method-specific type code>",
  "maxRowsInMemory": 1000000,
  "maxBytesInMemory": <one-sixth of JVM memory>,
  "indexSpec": {
    "bitmap": { "type": "roaring" },
    "dimensionCompression": "lz4",
    "metricCompression": "lz4",
    "longEncoding": "longs"
  },
  <other ingestion-method-specific properties>
}

Field	Description	Default
type	Each ingestion method has its own tuning type code. You must specify the type code that matches your ingestion method. Common options are `index`, `hadoop`, `kafka`, and `kinesis`.
maxRowsInMemory	The maximum number of records to store in memory before persisting to disk. Note that this is the number of rows post-rollup, and so it may not be equal to the number of input records. Ingested records will be persisted to disk when either `maxRowsInMemory` or `maxBytesInMemory` are reached (whichever happens first).	`1000000`
maxBytesInMemory	The maximum aggregate size of records, in bytes, to store in the JVM heap before persisting. This is based on a rough estimate of memory usage. Ingested records will be persisted to disk when either `maxRowsInMemory` or `maxBytesInMemory` are reached (whichever happens first). `maxBytesInMemory` also includes heap usage of artifacts created from intermediary persists. This means that after every persist, the amount of `maxBytesInMemory` until next persist will decreases, and task will fail when the sum of bytes of all intermediary persisted artifacts exceeds `maxBytesInMemory`. Setting maxBytesInMemory to -1 disables this check, meaning Druid will rely entirely on maxRowsInMemory to control memory usage. Setting it to zero means the default value will be used (one-sixth of JVM heap size). Note that the estimate of memory usage is designed to be an overestimate, and can be especially high when using complex ingest-time aggregators, including sketches. If this causes your indexing workloads to persist to disk too often, you can set maxBytesInMemory to -1 and rely on maxRowsInMemory instead.	One-sixth of max JVM heap size
skipBytesInMemoryOverheadCheck	The calculation of maxBytesInMemory takes into account overhead objects created during ingestion and each intermediate persist. Setting this to true can exclude the bytes of these overhead objects from maxBytesInMemory check.	false
indexSpec	Tune how data is indexed. See below for more information.	See table below
Other properties	Each ingestion method has its own list of additional tuning properties. See the documentation for each method for a full list: Kafka indexing service, Kinesis indexing service, Native batch, and Hadoop-based.

`indexSpec`

The indexSpec object can include the following properties:

Field	Description	Default
bitmap	Compression format for bitmap indexes. Should be a JSON object with `type` set to `roaring` or `concise`. For type `roaring`, the boolean property `compressRunOnSerialization` (defaults to true) controls whether or not run-length encoding will be used when it is determined to be more space-efficient.	`{"type": "roaring"}`
dimensionCompression	Compression format for dimension columns. Options are `lz4`, `lzf`, or `uncompressed`.	`lz4`
metricCompression	Compression format for primitive type metric columns. Options are `lz4`, `lzf`, `uncompressed`, or `none` (which is more efficient than `uncompressed`, but not supported by older versions of Druid).	`lz4`
longEncoding	Encoding format for long-typed columns. Applies regardless of whether they are dimensions or metrics. Options are `auto` or `longs`. `auto` encodes the values using offset or lookup table depending on column cardinality, and store them with variable size. `longs` stores the value as-is with 8 bytes each.	`longs`

Beyond these properties, each ingestion method has its own specific tuning properties. See the documentation for each ingestion method for details.

27 KiB Raw Blame History

dataSchema

dataSource

timestampSpec

dimensionsSpec

Dimension objects

Inclusions and exclusions

metricsSpec

granularitySpec

transformSpec

Transforms

Filter

Legacy dataSchema spec

parser (Deprecated)

flattenSpec

ioConfig

tuningConfig

indexSpec

27 KiB

Raw Blame History

`dataSchema`

`dataSource`

`timestampSpec`

`dimensionsSpec`

`metricsSpec`

`granularitySpec`

`transformSpec`

Legacy `dataSchema` spec

`parser` (Deprecated)

`flattenSpec`

`ioConfig`

`tuningConfig`

`indexSpec`