druid-docs-cn/ingestion/index.md

76 KiB
Raw Blame History

数据导入

All data in Druid is organized into segments, which are data files each of which may have up to a few million rows. Loading data in Druid is called ingestion or indexing, and consists of reading data from a source system and creating segments based on that data.

In most ingestion methods, the Druid MiddleManager processes (or the Indexer processes) load your source data. One exception is Hadoop-based ingestion, where this work is instead done using a Hadoop MapReduce job on YARN (although MiddleManager or Indexer processes are still involved in starting and monitoring the Hadoop jobs).

Once segments have been generated and stored in deep storage, they are loaded by Historical processes. For more details on how this works, see the Storage design section of Druid's design documentation.

How to use this documentation

This page you are currently reading provides information about universal Druid ingestion concepts, and about configurations that are common to all ingestion methods.

The individual pages for each ingestion method provide additional information about concepts and configurations that are unique to each ingestion method.

We recommend reading (or at least skimming) this universal page first, and then referring to the page for the ingestion method or methods that you have chosen.

Ingestion methods

The table below lists Druid's most common data ingestion methods, along with comparisons to help you choose the best one for your situation. Each ingestion method supports its own set of source systems to pull from. For details about how each method works, as well as configuration properties specific to that method, check out its documentation page.

Streaming

The most recommended, and most popular, method of streaming ingestion is the Kafka indexing service that reads directly from Kafka. Alternatively, the Kinesis indexing service works with Amazon Kinesis Data Streams.

This table compares the options:

Method Kafka Kinesis
Supervisor type kafka kinesis
How it works Druid reads directly from Apache Kafka. Druid reads directly from Amazon Kinesis.
Can ingest late data? Yes Yes
Exactly-once guarantees? Yes Yes

Batch

When doing batch loads from files, you should use one-time tasks, and you have three options: index_parallel (native batch; parallel), index_hadoop (Hadoop-based), or index (native batch; single-task).

In general, we recommend native batch whenever it meets your needs, since the setup is simpler (it does not depend on an external Hadoop cluster). However, there are still scenarios where Hadoop-based batch ingestion might be a better choice, for example when you already have a running Hadoop cluster and want to use the cluster resource of the existing cluster for batch ingestion.

This table compares the three available options:

Method Native batch (parallel) Hadoop-based Native batch (simple)
Task type index_parallel index_hadoop index
Parallel? Yes, if inputFormat is splittable and maxNumConcurrentSubTasks > 1 in tuningConfig. See data format documentation for details. Yes, always. No. Each task is single-threaded.
Can append or overwrite? Yes, both. Overwrite only. Yes, both.
External dependencies None. Hadoop cluster (Druid submits Map/Reduce jobs). None.
Input locations Any inputSource. Any Hadoop FileSystem or Druid datasource. Any inputSource.
File formats Any inputFormat. Any Hadoop InputFormat. Any inputFormat.
Rollup modes Perfect if forceGuaranteedRollup = true in the tuningConfig. Always perfect. Perfect if forceGuaranteedRollup = true in the tuningConfig.
Partitioning options Dynamic, hash-based, and range-based partitioning methods are available. See Partitions Spec for details. Hash-based or range-based partitioning via partitionsSpec. Dynamic and hash-based partitioning methods are available. See Partitions Spec for details.

Druid's data model

Datasources

Druid data is stored in datasources, which are similar to tables in a traditional RDBMS. Druid offers a unique data modeling system that bears similarity to both relational and timeseries models.

Primary timestamp

Druid schemas must always include a primary timestamp. The primary timestamp is used for partitioning and sorting your data. Druid queries are able to rapidly identify and retrieve data corresponding to time ranges of the primary timestamp column. Druid is also able to use the primary timestamp column for time-based data management operations such as dropping time chunks, overwriting time chunks, and time-based retention rules.

The primary timestamp is parsed based on the timestampSpec. In addition, the granularitySpec controls other important operations that are based on the primary timestamp. Regardless of which input field the primary timestamp is read from, it will always be stored as a column named __time in your Druid datasource.

If you have more than one timestamp column, you can store the others as secondary timestamps.

Dimensions

Dimensions are columns that are stored as-is and can be used for any purpose. You can group, filter, or apply aggregators to dimensions at query time in an ad-hoc manner. If you run with rollup disabled, then the set of dimensions is simply treated like a set of columns to ingest, and behaves exactly as you would expect from a typical database that does not support a rollup feature.

Dimensions are configured through the dimensionsSpec.

Metrics

Metrics are columns that are stored in an aggregated form. They are most useful when rollup is enabled. Specifying a metric allows you to choose an aggregation function for Druid to apply to each row during ingestion. This has two benefits:

  1. If rollup is enabled, multiple rows can be collapsed into one row even while retaining summary information. In the rollup tutorial, this is used to collapse netflow data to a single row per (minute, srcIP, dstIP) tuple, while retaining aggregate information about total packet and byte counts.
  2. Some aggregators, especially approximate ones, can be computed faster at query time even on non-rolled-up data if they are partially computed at ingestion time.

Metrics are configured through the metricsSpec.

Rollup

What is rollup?

Druid can roll up data as it is ingested to minimize the amount of raw data that needs to be stored. Rollup is a form of summarization or pre-aggregation. In practice, rolling up data can dramatically reduce the size of data that needs to be stored, reducing row counts by potentially orders of magnitude. This storage reduction does come at a cost: as we roll up data, we lose the ability to query individual events.

When rollup is disabled, Druid loads each row as-is without doing any form of pre-aggregation. This mode is similar to what you would expect from a typical database that does not support a rollup feature.

When rollup is enabled, then any rows that have identical dimensions and timestamp to each other (after queryGranularity-based truncation) can be collapsed, or rolled up, into a single row in Druid.

By default, rollup is enabled.

Enabling or disabling rollup

Rollup is controlled by the rollup setting in the granularitySpec. By default, it is true (enabled). Set this to false if you want Druid to store each record as-is, without any rollup summarization.

Example of rollup

For an example of how to configure rollup, and of how the feature will modify your data, check out the rollup tutorial.

Maximizing rollup ratio

You can measure the rollup ratio of a datasource by comparing the number of rows in Druid (COUNT) with the number of ingested events. One way to do this is with a Druid SQL query such as the following, where "count" refers to a count-type metric generated at ingestion time:

SELECT SUM("count") / (COUNT(*) * 1.0)
FROM datasource

The higher this number is, the more benefit you are gaining from rollup.

See Counting the number of ingested events on the "Schema design" page for more details about how counting works when rollup is enabled.

Tips for maximizing rollup:

  • Generally, the fewer dimensions you have, and the lower the cardinality of your dimensions, the better rollup ratios you will achieve.
  • Use sketches to avoid storing high cardinality dimensions, which harm rollup ratios.
  • Adjusting queryGranularity at ingestion time (for example, using PT5M instead of PT1M) increases the likelihood of two rows in Druid having matching timestamps, and can improve your rollup ratios.
  • It can be beneficial to load the same data into more than one Druid datasource. Some users choose to create a "full" datasource that has rollup disabled (or enabled, but with a minimal rollup ratio) and an "abbreviated" datasource that has fewer dimensions and a higher rollup ratio. When queries only involve dimensions in the "abbreviated" set, using that datasource leads to much faster query times. This can often be done with just a small increase in storage footprint, since abbreviated datasources tend to be substantially smaller.
  • If you are using a best-effort rollup ingestion configuration that does not guarantee perfect rollup, you can potentially improve your rollup ratio by switching to a guaranteed perfect rollup option, or by reindexing or compacting your data in the background after initial ingestion.

Perfect rollup vs Best-effort rollup

Some Druid ingestion methods guarantee perfect rollup, meaning that input data are perfectly aggregated at ingestion time. Others offer best-effort rollup, meaning that input data might not be perfectly aggregated and thus there could be multiple segments holding rows with the same timestamp and dimension values.

In general, ingestion methods that offer best-effort rollup do this because they are either parallelizing ingestion without a shuffling step (which would be required for perfect rollup), or because they are finalizing and publishing segments before all data for a time chunk has been received, which we call incremental publishing. In both of these cases, records that could theoretically be rolled up may end up in different segments. All types of streaming ingestion run in this mode.

Ingestion methods that guarantee perfect rollup do it with an additional preprocessing step to determine intervals and partitioning before the actual data ingestion stage. This preprocessing step scans the entire input dataset, which generally increases the time required for ingestion, but provides information necessary for perfect rollup.

The following table shows how each method handles rollup:

Method How it works
Native batch index_parallel and index type may be either perfect or best-effort, based on configuration.
Hadoop Always perfect.
Kafka indexing service Always best-effort.
Kinesis indexing service Always best-effort.

Partitioning

Why partition?

Optimal partitioning and sorting of segments within your datasources can have substantial impact on footprint and performance.

Druid datasources are always partitioned by time into time chunks, and each time chunk contains one or more segments. This partitioning happens for all ingestion methods, and is based on the segmentGranularity parameter of your ingestion spec's dataSchema.

The segments within a particular time chunk may also be partitioned further, using options that vary based on the ingestion type you have chosen. In general, doing this secondary partitioning using a particular dimension will improve locality, meaning that rows with the same value for that dimension are stored together and can be accessed quickly.

You will usually get the best performance and smallest overall footprint by partitioning your data on some "natural" dimension that you often filter by, if one exists. This will often improve compression - users have reported threefold storage size decreases - and it also tends to improve query performance as well.

Partitioning and sorting are best friends! If you do have a "natural" partitioning dimension, you should also consider placing it first in the dimensions list of your dimensionsSpec, which tells Druid to sort rows within each segment by that column. This will often improve compression even more, beyond the improvement gained by partitioning alone.

However, note that currently, Druid always sorts rows within a segment by timestamp first, even before the first dimension listed in your dimensionsSpec. This can prevent dimension sorting from being maximally effective. If necessary, you can work around this limitation by setting queryGranularity equal to segmentGranularity in your granularitySpec, which will set all timestamps within the segment to the same value, and by saving your "real" timestamp as a secondary timestamp. This limitation may be removed in a future version of Druid.

How to set up partitioning

Not all ingestion methods support an explicit partitioning configuration, and not all have equivalent levels of flexibility. As of current Druid versions, If you are doing initial ingestion through a less-flexible method (like Kafka) then you can use reindexing or compaction to repartition your data after it is initially ingested. This is a powerful technique: you can use it to ensure that any data older than a certain threshold is optimally partitioned, even as you continuously add new data from a stream.

The following table shows how each ingestion method handles partitioning:

Method How it works
Native batch Configured using partitionsSpec inside the tuningConfig.
Hadoop Configured using partitionsSpec inside the tuningConfig.
Kafka indexing service Partitioning in Druid is guided by how your Kafka topic is partitioned. You can also reindex or compact to repartition after initial ingestion.
Kinesis indexing service Partitioning in Druid is guided by how your Kinesis stream is sharded. You can also reindex or compact to repartition after initial ingestion.

Note that, of course, one way to partition data is to load it into separate datasources. This is a perfectly viable approach and works very well when the number of datasources does not lead to excessive per-datasource overheads. If you go with this approach, then you can ignore this section, since it is describing how to set up partitioning within a single datasource.

For more details on splitting data up into separate datasources, and potential operational considerations, refer to the Multitenancy considerations page.

Ingestion specs

No matter what ingestion method you use, data is loaded into Druid using either one-time tasks or ongoing "supervisors" (which run and supervise a set of tasks over time). In any case, part of the task or supervisor definition is an ingestion spec.

Ingestion specs consists of three main components:

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:

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 dataSchemadataSource 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 dataSchematimestampSpec 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 dataSchemadimensionsSpec 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:

  1. 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.
  2. 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.
  3. Any field listed in dimensionExclusions is excluded.
  4. The field listed as column in the timestampSpec is excluded.
  5. Any field used as an input to an aggregator from the metricsSpec is excluded.
  6. Any field with the same name as an aggregator from the metricsSpec is excluded.
  7. 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 dataSchemametricsSpec 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 dataSchemagranularitySpec and is responsible for configuring the following operations:

  1. Partitioning a datasource into time chunks (via segmentGranularity).
  2. Truncating the timestamp, if desired (via queryGranularity).
  3. Specifying which time chunks of segments should be created, for batch ingestion (via intervals).
  4. 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 describing what time chunks of segments should be created. If type is set to uniform, this list will be broken up and rounded-off based on the segmentGranularity. If type is set to arbitrary, this list will be used as-is.

If null or not provided, batch ingestion tasks will generally determine which time chunks to output based on what timestamps are 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 will 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 dataSchematransformSpec 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.

parser (Deprecated)

In legacy dataSchema, the parser is located in the dataSchemaparser 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:

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 dataSchemaparserparseSpecflattenSpec 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": "concise"}
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.

数据摄入

综述

Druid中的所有数据都被组织成这些段是数据文件通常每个段最多有几百万行。在Druid中加载数据称为摄取或索引,它包括从源系统读取数据并基于该数据创建段。

在大多数摄取方法中加载数据的工作由Druid MiddleManager 进程(或 Indexer 进程完成。一个例外是基于Hadoop的摄取这项工作是使用Hadoop MapReduce作业在YARN上完成的尽管MiddleManager或Indexer进程仍然参与启动和监视Hadoop作业。一旦段被生成并存储在 深层存储它们将被Historical进程加载。有关如何在引擎下工作的更多细节请参阅Druid设计文档的存储设计 部分。

如何使用本文档

当前正在阅读的这个页面提供了通用Druid摄取概念的信息以及 所有摄取方法 通用的配置信息。

每个摄取方法的单独页面提供了有关每个摄取方法独有的概念和配置的附加信息。

我们建议您先阅读(或至少略读)这个通用页面,然后参考您选择的一种或多种摄取方法的页面。

摄入方式

下表列出了Druid最常用的数据摄取方法帮助您根据自己的情况选择最佳方法。每个摄取方法都支持自己的一组源系统。有关每个方法如何工作的详细信息以及特定于该方法的配置属性请查看其文档页。

流式摄取

最推荐、也是最流行的流式摄取方法是直接从Kafka读取数据的 Kafka索引服务 。如果你喜欢KinesisKinesis索引服务 也能很好地工作。

下表比较了主要可用选项:

Method Kafka Kinesis Tranquility
Supervisor类型 kafka kinesis N/A
如何工作 Druid直接从 Apache Kafka读取数据 Druid直接从Amazon Kinesis中读取数据 Tranquility, 一个独立于Druid的库用来将数据推送到Druid
可以摄入迟到的数据 Yes Yes No(迟到的数据将会被基于 windowPeriod 的配置丢弃掉)
保证不重不丢Exactly-once Yes Yes No

批量摄取

从文件进行批加载时,应使用一次性 任务,并且有三个选项:index_parallel(本地并行批任务)、index_hadoop基于hadoopindex(本地简单批任务)。

一般来说如果本地批处理能满足您的需要时我们建议使用它因为设置更简单它不依赖于外部Hadoop集群。但是仍有一些情况下基于Hadoop的批摄取可能是更好的选择例如当您已经有一个正在运行的Hadoop集群并且希望使用现有集群的集群资源进行批摄取时。

此表比较了三个可用选项:

方式 本地批任务(并行) 基于Hadoop 本地批任务(简单)
任务类型 index_parallel index_hadoop index
并行? 如果 inputFormat 是可分割的且 tuningConfig 中的 maxNumConcurrentSubTasks > 1, 则 Yes Yes No每个任务都是单线程的
支持追加或者覆盖 都支持 只支持覆盖 都支持
外部依赖 Hadoop集群用来提交Map-Reduce任务
输入位置 任何 输入数据源 任何Hadoop文件系统或者Druid数据源 任何 输入数据源
文件格式 任何 输入格式 任何Hadoop输入格式 任何 输入格式
Rollup modes 如果 tuningConfig 中的 forceGuaranteedRollup = true, 则为 Perfect(最佳rollup) 总是Perfect最佳rollup 如果 tuningConfig 中的 forceGuaranteedRollup = true, 则为 Perfect(最佳rollup)
分区选项 可选的有Dynamic, hash-basedrange-based 三种分区方式,详情参见 分区规范 通过 partitionsSpec中指定 hash-basedrange-based分区 可选的有Dynamichash-based二种分区方式,详情参见 分区规范

Druid数据模型

数据源

Druid数据存储在数据源中与传统RDBMS中的表类似。Druid提供了一个独特的数据建模系统它与关系模型和时间序列模型都具有相似性。

主时间戳列

Druid Schema必须始终包含一个主时间戳。主时间戳用于对 数据进行分区和排序。Druid查询能够快速识别和检索与主时间戳列的时间范围相对应的数据。Druid还可以将主时间戳列用于基于时间的数据管理操作,例如删除时间块、覆盖时间块和基于时间的保留规则。

主时间戳基于 timestampSpec 进行解析。此外,granularitySpec 控制基于主时间戳的其他重要操作。无论从哪个输入字段读取主时间戳,它都将作为名为 __time 的列存储在Druid数据源中。

如果有多个时间戳列,则可以将其他列存储为 辅助时间戳

维度

维度是按原样存储的列,可以用于任何目的, 可以在查询时以特殊方式对维度进行分组、筛选或应用聚合器。如果在禁用了 rollup 的情况下运行那么该维度集将被简单地视为要摄取的一组列并且其行为与不支持rollup功能的典型数据库的预期完全相同。

通过 dimensionSpec 配置维度。

指标

Metrics是以聚合形式存储的列。启用 rollup它们最有用。指定一个Metric允许您为Druid选择一个聚合函数以便在摄取期间应用于每一行。这有两个好处

  1. 如果启用了 rollup,即使保留摘要信息,也可以将多行折叠为一行。在 Rollup教程这用于将netflow数据折叠为每minutesrcIPdstIP)元组一行,同时保留有关总数据包和字节计数的聚合信息。
  2. 一些聚合器,特别是近似聚合器,即使在非汇总数据上,如果在接收时部分计算,也可以在查询时更快地计算它们。

Metrics是通过 metricsSpec 配置的。

Rollup

什么是rollup

Druid可以在接收过程中将数据进行汇总以最小化需要存储的原始数据量。Rollup是一种汇总或预聚合的形式。实际上Rollup可以极大地减少需要存储的数据的大小从而潜在地减少行数的数量级。这种存储量的减少是有代价的当我们汇总数据时我们就失去了查询单个事件的能力。

禁用rollup时Druid将按原样加载每一行而不进行任何形式的预聚合。此模式类似于您对不支持汇总功能的典型数据库的期望。

如果启用了rollup那么任何具有相同维度时间戳的行(在基于 queryGranularity 的截断之后都可以在Druid中折叠或汇总为一行。

rollup默认是启用状态。

启用或者禁用rollup

Rollup由 granularitySpec 中的 rollup 配置项控制。 默认情况下,值为 true(启用状态)。如果你想让Druid按原样存储每条记录而不需要任何汇总将该值设置为 false

rollup示例

有关如何配置Rollup以及该特性将如何修改数据的示例请参阅Rollup教程

最大化rollup比率

通过比较Druid中的行数和接收的事件数可以测量数据源的汇总率。这个数字越高从汇总中获得的好处就越多。一种方法是使用Druid SQL查询,比如:

SELECT SUM("cnt") / COUNT(*) * 1.0 FROM datasource

在这个查询中,cnt 应该引用在摄取时指定的"count"类型Metrics。有关启用汇总时计数工作方式的详细信息请参阅"架构设计"页上的 计数接收事件数

最大化Rollup的提示

  • 一般来说,拥有的维度越少,维度的基数越低,您将获得更好的汇总比率
  • 使用 Sketches 避免存储高基数维度,因为会损害汇总比率
  • 在摄入时调整 queryGranularity(例如,使用 PT5M 而不是 PT1M 会增加Druid中两行具有匹配时间戳的可能性并可以提高汇总率
  • 将相同的数据加载到多个Druid数据源中是有益的。有些用户选择创建禁用汇总或启用汇总但汇总比率最小的"完整"数据源和具有较少维度和较高汇总比率的"缩写"数据源。当查询只涉及"缩写"集里边的维度时,使用该数据源将导致更快的查询时间,这种方案只需稍微增加存储空间即可完成,因为简化的数据源往往要小得多。
  • 如果您使用的 尽力而为的汇总(best-effort rollup) 摄取配置不能保证完全汇总(perfect rollup),则可以通过切换到保证的完全汇总选项,或在初始摄取后在后台重新编制(reindex)数据索引,潜在地提高汇总比率。

最佳rollup VS 尽可能rollup

一些Druid摄取方法保证了完美的汇总(perfect rollup),这意味着输入数据在摄取时被完美地聚合。另一些则提供了尽力而为的汇总(best-effort rollup),这意味着输入数据可能无法完全聚合,因此可能有多个段保存具有相同时间戳和维度值的行。

一般来说,提供尽力而为的汇总(best-effort rollup)的摄取方法之所以这样做,是因为它们要么是在没有清洗步骤(这是完美的汇总(perfect rollup)所必需的)的情况下并行摄取,要么是因为它们在接收到某个时间段的所有数据(我们称之为增量发布(incremental publishing))之前完成并发布段。在这两种情况下,理论上可以汇总的记录可能会以不同的段结束。所有类型的流接收都在此模式下运行。

保证*完美的汇总(perfect rollup)*的摄取方法通过额外的预处理步骤来确定实际数据摄取阶段之前的间隔和分区。此预处理步骤扫描整个输入数据集,这通常会增加摄取所需的时间,但提供完美汇总所需的信息。

下表显示了每个方法如何处理汇总:

方法 如何工作
本地批 基于配置,index_parallelindex 可以是完美的,也可以是最佳的。
Hadoop批 总是 perfect
Kafka索引服务 总是 best-effort
Kinesis索引服务 总是 best-effort

分区

为什么分区

数据源中段的最佳分区和排序会对占用空间和性能产生重大影响 。 Druid数据源总是按时间划分为时间块,每个时间块包含一个或多个段。此分区适用于所有摄取方法,并基于摄取规范的 dataSchema 中的 segmentGranularity参数。

特定时间块内的段也可以进一步分区,使用的选项根据您选择的摄取类型而不同。一般来说,使用特定维度执行此辅助分区将改善局部性,这意味着具有该维度相同值的行存储在一起,并且可以快速访问。

通常,通过将数据分区到一些常用来做过滤操作的维度(如果存在的话)上,可以获得最佳性能和最小的总体占用空间。而且,这种分区通常会改善压缩性能而且还往往会提高查询性能(用户报告存储容量减少了三倍)。

[!WARNING] 分区和排序是最好的朋友!如果您确实有一个天然的分区维度,那么您还应该考虑将它放在 dimensionsSpecdimension 列表中的第一个维度它告诉Druid按照该列对每个段中的行进行排序。除了单独分区所获得的改进之外这通常还会进一步改进压缩。 但是请注意目前Druid总是首先按时间戳对一个段内的行进行排序甚至在 dimensionsSpec 中列出的第一个维度之前,这将使得维度排序达不到最大效率。如果需要,可以通过在 granularitySpec 中将 queryGranularity 设置为等于 segmentGranularity 的值来解决此限制,这将把段内的所有时间戳设置为相同的值,并将"真实"时间戳保存为辅助时间戳。这个限制可能在Druid的未来版本中被移除。

如何设置分区

并不是所有的摄入方式都支持显式的分区配置也不是所有的方法都具有同样的灵活性。在当前的Druid版本中如果您是通过一个不太灵活的方法如Kafka进行初始摄取那么您可以使用 重新索引的技术(reindex),在最初摄取数据后对其重新分区。这是一种强大的技术:即使您不断地从流中添加新数据, 也可以使用它来确保任何早于某个阈值的数据都得到最佳分区。

下表显示了每个摄取方法如何处理分区:

方法 如何工作
本地批 通过 tuningConfig 中的 partitionsSpec
Hadoop批 通过 tuningConfig 中的 partitionsSpec
Kafka索引服务 Druid中的分区是由Kafka主题的分区方式决定的。您可以在初次摄入后 重新索引的技术(reindex)以重新分区
Kinesis索引服务 Druid中的分区是由Kinesis流的分区方式决定的。您可以在初次摄入后 重新索引的技术(reindex)以重新分区

[!WARNING]

注意,当然,划分数据的一种方法是将其加载到分开的数据源中。这是一种完全可行的方法,当数据源的数量不会导致每个数据源的开销过大时,它可以很好地工作。如果使用这种方法,那么可以忽略这一部分,因为这部分描述了如何在单个数据源中设置分区。

有关将数据拆分为单独数据源的详细信息以及潜在的操作注意事项,请参阅 多租户注意事项

摄入规范

无论使用哪一种摄入方式,数据要么是通过一次性tasks或者通过持续性的"supervisor"(运行并监控一段时间内的一系列任务)来被加载到Druid中。 在任一种情况下task或者supervisor的定义都在摄入规范中定义。

摄入规范包括以下三个主要的部分:

一个 index_parallel 类型任务的示例摄入规范如下:

{
  "type": "index_parallel",
  "spec": {
    "dataSchema": {
      "dataSource": "wikipedia",
      "timestampSpec": {
        "column": "timestamp",
        "format": "auto"
      },
      "dimensionsSpec": {
        "dimensions": [
          { "type": "string", "page" },
          { "type": "string", "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"
    }
  }
}

该部分中支持的特定选项依赖于选择的摄入方法。 更多的示例,可以参考每一种摄入方法的文档。

您还可以不用编写一个摄入规范,可视化的加载数据,该功能位于 Druid控制台 的 "Load Data" 视图中。 Druid可视化数据加载器目前支持 Kafka, Kinesis本地批 模式。

dataSchema

[!WARNING]

dataSchema 规范在0.17.0版本中做了更改,新的规范支持除Hadoop摄取方式外的所有方式。 可以在 过时的 dataSchema 规范查看老的规范

dataSchema 包含了以下部分:

一个 dataSchema 如下:

"dataSchema": {
  "dataSource": "wikipedia",
  "timestampSpec": {
    "column": "timestamp",
    "format": "auto"
  },
  "dimensionsSpec": {
    "dimensions": [
      { "type": "string", "page" },
      { "type": "string", "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

dataSource 位于 dataSchema -> dataSource 中,简单的标识了数据将被写入的数据源的名称,示例如下:

"dataSource": "my-first-datasource"
timestampSpec

timestampSpec 位于 dataSchema -> timestampSpec 中,用来配置 主时间戳, 示例如下:

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

[!WARNING] 概念上输入数据被读取后Druid会以一个特定的顺序来对数据应用摄入规范 首先 flattenSpec(如果有),然后 timestampSpec, 然后 transformSpec ,最后是 dimensionsSpecmetricsSpec。在编写摄入规范时需要牢记这一点

timestampSpec 可以包含以下的部分:

字段 描述 默认值
column 要从中读取主时间戳的输入行字段。

不管这个输入字段的名称是什么,主时间戳总是作为一个名为"__time"的列存储在您的Druid数据源中
timestamp
format 时间戳格式,可选项有:
  • iso: 使用"T"分割的ISO8601像"2000-01-01T01:02:03.456"
  • posix: 自纪元以来的秒数
  • millis: 自纪元以来的毫秒数
  • micro: 自纪元以来的微秒数
  • nano: 自纪元以来的纳秒数
  • auto: 自动检测ISO或者毫秒格式
  • 任何 Joda DateTimeFormat字符串
auto
missingValue 用于具有空或缺少时间戳列的输入记录的时间戳。应该是ISO8601格式"2000-01-01T01:02:03.456"。由于Druid需要一个主时间戳因此此设置对于接收根本没有任何时间戳的数据集非常有用。 none
dimensionSpec

dimensionsSpec 位于 dataSchema -> dimensionsSpec, 用来配置维度。示例如下:

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

[!WARNING] 概念上输入数据被读取后Druid会以一个特定的顺序来对数据应用摄入规范 首先 flattenSpec(如果有),然后 timestampSpec, 然后 transformSpec ,最后是 dimensionsSpecmetricsSpec。在编写摄入规范时需要牢记这一点

dimensionsSpec 可以包括以下部分:

字段 描述 默认值
dimensions 维度名称或者对象的列表,在 dimensionsdimensionExclusions 中不能包含相同的列。

如果该配置为一个空数组Druid将会把所有未出现在 dimensionExclusions 中的非时间、非指标列当做字符串类型的维度列,参见Inclusions and exclusions
[]
dimensionExclusions 在摄取中需要排除的列名称,在该配置中只支持名称,不支持对象。在 dimensionsdimensionExclusions 中不能包含相同的列。 []
spatialDimensions 一个空间维度的数组 []
Dimension objects

dimensions 列的每一个维度可以是一个名称,也可以是一个对象。 提供一个名称等价于提供了一个给定名称的 string 类型的维度对象。例如: page 等价于 {"name": "page", "type": "string"}

维度对象可以有以下的部分:

字段 描述 默认值
type string, long, float 或者 double string
name 维度名称,将用作从输入记录中读取的字段名,以及存储在生成的段中的列名。

注意: 如果想在摄取的时候重新命名列,可以使用 transformSpec
none必填
createBitmapIndex 对于字符串类型的维度,是否应为生成的段中的列创建位图索引。创建位图索引需要更多存储空间,但会加快某些类型的筛选(特别是相等和前缀筛选)。仅支持字符串类型的维度。 true
Inclusions and exclusions

Druid以两种可能的方式来解释 dimensionsSpec : normalschemaless

dimensions 或者 spatialDimensions 为非空时, 将会采用正常的解释方式。 在该情况下, 前边说的两个列表结合起来的集合当做摄入的维度集合。

dimensionsspatialDimensions 同时为空或者null时候将会采用无模式的解释方式。 在该情况下,维度集合由以下方式决定:

  1. 首先,从 inputFormat (或者 flattenSpec, 如果正在使用 )中所有输入字段集合开始
  2. 排除掉任何在 dimensionExclusions 中的列
  3. 排除掉在 timestampSpec 中的时间列
  4. 排除掉 metricsSpec 中用于聚合器输入的列
  5. 排除掉 metricsSpec 中任何与聚合器同名的列
  6. 所有的其他字段都被按照默认配置摄入为 string 类型的维度

[!WARNING] 注意:在无模式的维度解释方式中,由 transformSpec 生成的列当前并未考虑。

metricsSpec

metricsSpec 位于 dataSchema -> metricsSpec 中,是一个在摄入阶段要应用的 聚合器 列表。 在启用了 rollup 时是很有用的,因为它将配置如何在摄入阶段进行聚合。

一个 metricsSpec 实例如下:

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

[!WARNING] 通常,当 rollup 被禁用时,应该有一个空的 metricsSpec因为没有rollupDruid不会在摄取时进行任何的聚合所以没有理由包含摄取时聚合器。但是在某些情况下定义Metrics仍然是有意义的例如如果要创建一个复杂的列作为 近似聚合 的预计算部分,则只能通过在 metricsSpec 中定义度量来实现

granularitySpec

granularitySpec 位于 dataSchema -> granularitySpec, 用来配置以下操作:

  1. 通过 segmentGranularity 来将数据源分区到 时间块
  2. 如果需要的话,通过 queryGranularity 来截断时间戳
  3. 通过 interval 来指定批摄取中应创建段的时间块
  4. 通过 rollup 来指定是否在摄取时进行汇总

除了 rollup, 这些操作都是基于 主时间戳列

一个 granularitySpec 实例如下:

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

granularitySpec 可以有以下的部分:

字段 描述 默认值
type uniform 或者 arbitrary ,大多数时候使用 uniform uniform
segmentGranularity 数据源的 时间分块 粒度。每个时间块可以创建多个段, 例如,当设置为 day 时,同一天的事件属于同一时间块,该时间块可以根据其他配置和输入大小进一步划分为多个段。这里可以提供任何粒度。请注意,同一时间块中的所有段应具有相同的段粒度。

如果 type 字段设置为 arbitrary 则忽略
day
queryGranularity 每个段内时间戳存储的分辨率, 必须等于或比 segmentGranularity 更细。这将是您可以查询的最细粒度,并且仍然可以查询到合理的结果。但是请注意,您仍然可以在比此粒度更粗的场景进行查询,例如 "minute"的值意味着记录将以分钟的粒度存储并且可以在分钟的任意倍数包括分钟、5分钟、小时等进行查询。

这里可以提供任何 粒度 。使用 none 按原样存储时间戳,而不进行任何截断。请注意,即使将 queryGranularity 设置为 none,也将应用 rollup
none
rollup 是否在摄取时使用 rollup。 注意:即使 queryGranularity 设置为 nonerollup也仍然是有效的当数据具有相同的时间戳时数据将被汇总 true
interval 描述应该创建段的时间块的间隔列表。如果 type 设置为uniform,则此列表将根据 segmentGranularity 进行拆分和舍入。如果 type 设置为 arbitrary ,则将按原样使用此列表。

如果该值不提供或者为空值,则批处理摄取任务通常会根据在输入数据中找到的时间戳来确定要输出的时间块。

如果指定,批处理摄取任务可以跳过确定分区阶段,这可能会导致更快的摄取。批量摄取任务也可以预先请求它们的所有锁,而不是逐个请求。批处理摄取任务将丢弃任何时间戳超出指定间隔的记录。

在任何形式的流摄取中忽略该配置。
null
transformSpec

transformSpec 位于 dataSchema -> transformSpec,用来摄取时转换和过滤输入数据。 一个 transformSpec 实例如下:

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

[!WARNING] 概念上输入数据被读取后Druid会以一个特定的顺序来对数据应用摄入规范 首先 flattenSpec(如果有),然后 timestampSpec, 然后 transformSpec ,最后是 dimensionsSpecmetricsSpec。在编写摄入规范时需要牢记这一点

过时的 dataSchema 规范

[!WARNING]

dataSchema 规范在0.17.0版本中做了更改,新的规范支持除Hadoop摄取方式外的所有方式。 可以在 dataSchema查看老的规范

除了上面 dataSchema 一节中列出的组件之外,过时的 dataSchema 规范还有以下两个组件。

  • input row parser, flatten of nested data

parser(已废弃) 在过时的 dataSchema 中,parser 位于 dataSchema -> parser中,负责配置与解析输入记录相关的各种项。由于 parser 已经废弃,不推荐使用,强烈建议改用 inputFormat。 对于 inputFormat 和支持的 parser 类型,可以参见 数据格式

parseSpec主要部分的详细,参见他们的子部分:

一个 parser 实例如下:

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

flattenSpec 在过时的 dataSchema 中,flattenSpec 位于dataSchema -> parser -> parseSpec -> flattenSpec负责在潜在的嵌套输入数据如JSON、Avro等和Druid的数据模型之间架起桥梁。有关详细信息请参见 flattenSpec

ioConfig

ioConfig 影响从源系统如Apache Kafka、Amazon S3、挂载的文件系统或任何其他受支持的源系统读取数据的方式。inputFormat 属性适用于除Hadoop摄取之外的所有摄取方法。Hadoop摄取仍然使用过时的 dataSchema 中的 [parser]。ioConfig 的其余部分特定于每个单独的摄取方法。读取JSON数据的 ioConfig 示例如下:

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

详情可以参见每个 摄取方式 提供的文档。

tuningConfig

优化属性在 tuningConfig 中指定,tuningConfig 位于摄取规范的顶层。有些属性适用于所有摄取方法,但大多数属性特定于每个单独的摄取方法。tuningConfig 将所有共享的公共属性设置为默认值的示例如下:

"tuningConfig": {
  "type": "<ingestion-method-specific type code>",
  "maxRowsInMemory": 1000000,
  "maxBytesInMemory": <one-sixth of JVM memory>,
  "indexSpec": {
    "bitmap": { "type": "concise" },
    "dimensionCompression": "lz4",
    "metricCompression": "lz4",
    "longEncoding": "longs"
  },
  <other ingestion-method-specific properties>
}
字段 描述 默认值
type 每一种摄入方式都有自己的类型,必须指定为与摄入方式匹配的类型。通常的选项有 index, hadoop, kafkakinesis
maxRowsInMemory 数据持久化到硬盘前在内存中存储的最大数据条数。 注意,这个数字是汇总后的,所以可能并不等于输入的记录数。 当摄入的数据达到 maxRowsInMemory 或者 maxBytesInMemory 时数据将被持久化到硬盘。 1000000
maxBytesInMemory 在持久化之前要存储在JVM堆中的数据最大字节数。这是基于对内存使用的粗略估计。当达到 maxRowsInMemorymaxBytesInMemory 时(以先发生的为准),摄取的记录将被持久化到磁盘。

maxBytesInMemory 设置为-1将禁用此检查这意味着Druid将完全依赖 maxRowsInMemory 来控制内存使用。将其设置为零意味着将使用默认值JVM堆大小的六分之一

请注意内存使用量的估计值被设计为高估值并且在使用复杂的摄取时聚合器包括sketches时可能特别高。如果这导致索引工作负载过于频繁地持久化到磁盘则可以将 maxBytesInMemory 设置为-1并转而依赖 maxRowsInMemory
JVM堆内存最大值的1/6
indexSpec 优化数据如何被索引,详情可以看下面的表格 看下面的表格
其他属性 每一种摄入方式都有其自己的优化属性。 详情可以查看每一种方法的文档。 Kafka索引服务, Kinesis索引服务, 本地批Hadoop批

indexSpec

上边表格中的 indexSpec 部分可以包含以下属性:

字段 描述 默认值
bitmap 位图索引的压缩格式。 需要一个 type 设置为 concise 或者 roaring 的JSON对象。对于 roaring类型,布尔属性compressRunOnSerialization默认为true控制在确定运行长度编码更节省空间时是否使用该编码。 {"type":"concise"}
dimensionCompression 维度列的压缩格式。 可选项有 lz4, lzf 或者 uncompressed lz4
metricCompression Metrics列的压缩格式。可选项有 lz4, lzf, uncompressed 或者 none(noneuncompressed 更有效但是在老版本的Druid不支持) lz4
longEncoding long类型列的编码格式。无论它们是维度还是Metrics都适用选项是 autolongauto 根据列基数使用偏移量或查找表对值进行编码,并以可变大小存储它们。longs 按原样存储值每个值8字节。 longs

除了这些属性之外,每个摄取方法都有自己的特定调整属性。有关详细信息,请参阅每个 摄取方法 的文档。