| <!DOCTYPE html><html lang="en"><head><meta charSet="utf-8"/><meta http-equiv="X-UA-Compatible" content="IE=edge"/><title>SQL-based ingestion concepts ยท Apache Druid</title><meta name="viewport" content="width=device-width, initial-scale=1.0"/><link rel="canonical" href="https://druid.apache.org/docs/26.0.0/multi-stage-query/concepts.html"/><meta name="generator" content="Docusaurus"/><meta name="description" content="<!--"/><meta name="docsearch:language" content="en"/><meta name="docsearch:version" content="26.0.0" /><meta property="og:title" content="SQL-based ingestion concepts ยท Apache Druid"/><meta property="og:type" content="website"/><meta property="og:url" content="https://druid.apache.org/index.html"/><meta property="og:description" content="<!--"/><meta property="og:image" content="https://druid.apache.org/img/druid_nav.png"/><meta name="twitter:card" content="summary"/><meta name="twitter:image" content="https://druid.apache.org/img/druid_nav.png"/><link rel="shortcut icon" href="/img/favicon.png"/><link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/docsearch.js@2/dist/cdn/docsearch.min.css"/><link rel="stylesheet" href="//cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/styles/default.min.css"/><script async="" src="https://www.googletagmanager.com/gtag/js?id=UA-131010415-1"></script><script> |
| window.dataLayer = window.dataLayer || []; |
| function gtag(){dataLayer.push(arguments); } |
| gtag('js', new Date()); |
| gtag('config', 'UA-131010415-1'); |
| </script><link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.7.2/css/all.css"/><link rel="stylesheet" href="/css/code-block-buttons.css"/><script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.4/clipboard.min.js"></script><script type="text/javascript" src="/js/code-block-buttons.js"></script><script src="/js/scrollSpy.js"></script><link rel="stylesheet" href="/css/main.css"/><script src="/js/codetabs.js"></script></head><body class="sideNavVisible separateOnPageNav"><div class="fixedHeaderContainer"><div class="headerWrapper wrapper"><header><a href="/"><img class="logo" src="/img/druid_nav.png" alt="Apache Druid"/></a><div class="navigationWrapper navigationSlider"><nav class="slidingNav"><ul class="nav-site nav-site-internal"><li class=""><a href="/technology" target="_self">Technology</a></li><li class=""><a href="/use-cases" target="_self">Use Cases</a></li><li class=""><a href="/druid-powered" target="_self">Powered By</a></li><li class="siteNavGroupActive"><a href="/docs/26.0.0/design/index.html" target="_self">Docs</a></li><li class=""><a href="/community/" target="_self">Community</a></li><li class=""><a href="https://www.apache.org" target="_self">Apache</a></li><li class=""><a href="/downloads.html" target="_self">Download</a></li><li class="navSearchWrapper reactNavSearchWrapper"><input type="text" id="search_input_react" placeholder="Search" title="Search"/></li></ul></nav></div></header></div></div><div class="navPusher"><div class="docMainWrapper wrapper"><div class="docsNavContainer" id="docsNav"><nav class="toc"><div class="toggleNav"><section class="navWrapper wrapper"><div class="navBreadcrumb wrapper"><div class="navToggle" id="navToggler"><div class="hamburger-menu"><div class="line1"></div><div class="line2"></div><div class="line3"></div></div></div><h2><i>โบ</i><span>SQL-based ingestion ๐</span></h2><div class="tocToggler" id="tocToggler"><i class="icon-toc"></i></div></div><div class="navGroups"><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Getting started<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/index.html">Introduction to Apache Druid</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/index.html">Quickstart (local)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/single-server.html">Single server deployment</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/cluster.html">Clustered deployment</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Tutorials<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-batch.html">Load files natively</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-msq-extern.html">Load files using SQL ๐</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-kafka.html">Load from Apache Kafka</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-batch-hadoop.html">Load from Apache Hadoop</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-query.html">Querying data</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-rollup.html">Roll-up</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-sketches-theta.html">Theta sketches</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-retention.html">Configuring data retention</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-update-data.html">Updating existing data</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-compaction.html">Compacting segments</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-delete-data.html">Deleting data</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-ingestion-spec.html">Writing an ingestion spec</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-transform-spec.html">Transforming input data</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/docker.html">Tutorial: Run with Docker</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-kerberos-hadoop.html">Kerberized HDFS deep storage</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-msq-convert-spec.html">Convert ingestion spec to SQL</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-sql-query-view.html">Get to know Query view</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-unnest-arrays.html">Unnesting arrays</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-jupyter-index.html">Jupyter Notebook tutorials</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/tutorials/tutorial-jdbc.html">JDBC connector</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Design<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/architecture.html">Design</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/segments.html">Segments</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/processes.html">Processes and servers</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/dependencies/deep-storage.html">Deep storage</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/dependencies/metadata-storage.html">Metadata storage</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/dependencies/zookeeper.html">ZooKeeper</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Ingestion<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/index.html">Ingestion</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/data-formats.html">Data formats</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/data-model.html">Data model</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/rollup.html">Data rollup</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/partitioning.html">Partitioning</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/ingestion-spec.html">Ingestion spec</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/schema-design.html">Schema design tips</a></li><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Stream ingestion</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kafka-ingestion.html">Apache Kafka ingestion</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kafka-supervisor-reference.html">Apache Kafka supervisor</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kafka-supervisor-operations.html">Apache Kafka operations</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kinesis-ingestion.html">Amazon Kinesis</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Batch ingestion</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/native-batch.html">Native batch</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/native-batch-input-sources.html">Native batch: input sources</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/migrate-from-firehose.html">Migrate from firehose</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/hadoop.html">Hadoop-based</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">SQL-based ingestion ๐</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/index.html">Overview</a></li><li class="navListItem navListItemActive"><a class="navItem" href="/docs/26.0.0/multi-stage-query/concepts.html">Key concepts</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/api.html">API</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/security.html">Security</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/examples.html">Examples</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/reference.html">Reference</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/multi-stage-query/known-issues.html">Known issues</a></li></ul></div><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/tasks.html">Task reference</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/faq.html">Troubleshooting FAQ</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Data management<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/index.html">Overview</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/update.html">Data updates</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/delete.html">Data deletion</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/schema-changes.html">Schema changes</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/compaction.html">Compaction</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/data-management/automatic-compaction.html">Automatic compaction</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Querying<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Druid SQL</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql.html">Overview and syntax</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-data-types.html">SQL data types</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-operators.html">Operators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-scalar.html">Scalar functions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-aggregations.html">Aggregation functions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-multivalue-string-functions.html">Multi-value string functions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-json-functions.html">JSON functions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-functions.html">All functions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-api.html">Druid SQL API</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-jdbc.html">JDBC driver API</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-query-context.html">SQL query context</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-metadata-tables.html">SQL metadata tables</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sql-translation.html">SQL query translation</a></li></ul></div><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/querying.html">Native queries</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/query-execution.html">Query execution</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/troubleshooting.html">Troubleshooting</a></li><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Concepts</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/datasource.html">Datasources</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/joins.html">Joins</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/lookups.html">Lookups</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/multi-value-dimensions.html">Multi-value dimensions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/nested-columns.html">Nested columns</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/multitenancy.html">Multitenancy</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/caching.html">Query caching</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/using-caching.html">Using query caching</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/query-context.html">Query context</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Native query types</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/timeseriesquery.html">Timeseries</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/topnquery.html">TopN</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/groupbyquery.html">GroupBy</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/scan-query.html">Scan</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/searchquery.html">Search</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/timeboundaryquery.html">TimeBoundary</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/segmentmetadataquery.html">SegmentMetadata</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/datasourcemetadataquery.html">DatasourceMetadata</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Native query components</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/filters.html">Filters</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/granularities.html">Granularities</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/dimensionspecs.html">Dimensions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/aggregations.html">Aggregations</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/post-aggregations.html">Post-aggregations</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/misc/math-expr.html">Expressions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/having.html">Having filters (groupBy)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/limitspec.html">Sorting and limiting (groupBy)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/topnmetricspec.html">Sorting (topN)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/sorting-orders.html">String comparators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/virtual-columns.html">Virtual columns</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/geo.html">Spatial filters</a></li></ul></div></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Configuration<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/configuration/index.html">Configuration reference</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions.html">Extensions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/configuration/logging.html">Logging</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Operations<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/web-console.html">Web console</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/java.html">Java runtime</a></li><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Security</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/security-overview.html">Security overview</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/security-user-auth.html">User authentication and authorization</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/auth-ldap.html">LDAP auth</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/password-provider.html">Password providers</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/dynamic-config-provider.html">Dynamic Config Providers</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/tls-support.html">TLS support</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Performance tuning</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/basic-cluster-tuning.html">Basic cluster tuning</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/segment-optimization.html">Segment size optimization</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/mixed-workloads.html">Mixed workloads</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/http-compression.html">HTTP compression</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/clean-metadata-store.html">Automated metadata cleanup</a></li></ul></div><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Monitoring</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/request-logging.html">Request logging</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/metrics.html">Metrics</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/alerts.html">Alerts</a></li></ul></div><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/api-reference.html">API reference</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/high-availability.html">High availability</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/rolling-updates.html">Rolling updates</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/rule-configuration.html">Using rules to drop and retain data</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/other-hadoop.html">Working with different versions of Apache Hadoop</a></li><div class="navGroup subNavGroup"><h4 class="navGroupSubcategoryTitle">Misc</h4><ul><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/dump-segment.html">dump-segment tool</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/reset-cluster.html">reset-cluster tool</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/insert-segment-to-db.html">insert-segment-to-db tool</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/pull-deps.html">pull-deps tool</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/deep-storage-migration.html">Deep storage migration</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/export-metadata.html">Export Metadata Tool</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/metadata-migration.html">Metadata Migration</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/use_sbt_to_build_fat_jar.html">Content for build.sbt</a></li></ul></div></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Development<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/overview.html">Developing on Druid</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/modules.html">Creating extensions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/javascript.html">JavaScript functionality</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/build.html">Build from source</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/versioning.html">Versioning</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/experimental.html">Experimental features</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Misc<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/misc/papers-and-talks.html">Papers</a></li></ul></div><div class="navGroup"><h3 class="navGroupCategoryTitle collapsible">Hidden<span class="arrow"><svg width="24" height="24" viewBox="0 0 24 24"><path fill="#565656" d="M7.41 15.41L12 10.83l4.59 4.58L18 14l-6-6-6 6z"></path><path d="M0 0h24v24H0z" fill="none"></path></svg></span></h3><ul class="hide"><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-elasticsearch.html">Apache Druid vs Elasticsearch</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-key-value.html">Apache Druid vs. Key/Value Stores (HBase/Cassandra/OpenTSDB)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-kudu.html">Apache Druid vs Kudu</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-redshift.html">Apache Druid vs Redshift</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-spark.html">Apache Druid vs Spark</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/comparisons/druid-vs-sql-on-hadoop.html">Apache Druid vs SQL-on-Hadoop</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/auth.html">Authentication and Authorization</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/broker.html">Broker</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/coordinator.html">Coordinator Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/historical.html">Historical Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/indexer.html">Indexer Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/indexing-service.html">Indexing Service</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/middlemanager.html">MiddleManager Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/overlord.html">Overlord Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/router.html">Router Process</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/design/peons.html">Peons</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/approximate-histograms.html">Approximate Histogram aggregators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/avro.html">Apache Avro</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/azure.html">Microsoft Azure</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/bloom-filter.html">Bloom Filter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/datasketches-extension.html">DataSketches extension</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/datasketches-hll.html">DataSketches HLL Sketch module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/datasketches-quantiles.html">DataSketches Quantiles Sketch module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/datasketches-theta.html">DataSketches Theta Sketch module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/datasketches-tuple.html">DataSketches Tuple Sketch module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-basic-security.html">Basic Security</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-kerberos.html">Kerberos</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-lookups.html">Cached Lookup Module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-ranger-security.html">Apache Ranger Security</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/google.html">Google Cloud Storage</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/hdfs.html">HDFS</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kafka-extraction-namespace.html">Apache Kafka Lookups</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/lookups-cached-global.html">Globally Cached Lookups</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/mysql.html">MySQL Metadata Store</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/orc.html">ORC Extension</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-pac4j.html">Druid pac4j based Security extension</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/parquet.html">Apache Parquet Extension</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/postgresql.html">PostgreSQL Metadata Store</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/protobuf.html">Protobuf</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/s3.html">S3-compatible</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/simple-client-sslcontext.html">Simple SSLContext Provider Module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/stats.html">Stats aggregator</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/test-stats.html">Test Stats Aggregators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/druid-aws-rds.html">Druid AWS RDS Module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-core/kubernetes.html">Kubernetes</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/ambari-metrics-emitter.html">Ambari Metrics Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/cassandra.html">Apache Cassandra</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/cloudfiles.html">Rackspace Cloud Files</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/distinctcount.html">DistinctCount Aggregator</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/graphite.html">Graphite Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/influx.html">InfluxDB Line Protocol Parser</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/influxdb-emitter.html">InfluxDB Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/kafka-emitter.html">Kafka Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/materialized-view.html">Materialized View</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/momentsketch-quantiles.html">Moment Sketches for Approximate Quantiles module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/moving-average-query.html">Moving Average Query</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/opentsdb-emitter.html">OpenTSDB Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/redis-cache.html">Druid Redis Cache</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/sqlserver.html">Microsoft SQLServer</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/statsd.html">StatsD Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/tdigestsketch-quantiles.html">T-Digest Quantiles Sketch module</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/thrift.html">Thrift</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/time-min-max.html">Timestamp Min/Max aggregators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/gce-extensions.html">GCE Extensions</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/aliyun-oss.html">Aliyun OSS</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/development/extensions-contrib/prometheus.html">Prometheus Emitter</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/operations/kubernetes.html">kubernetes</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/hll-old.html">Cardinality/HyperUnique aggregators</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/querying/select-query.html">Select</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/native-batch-firehose.html">Firehose (deprecated)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/native-batch-simple-task.html">Native batch (simple)</a></li><li class="navListItem"><a class="navItem" href="/docs/26.0.0/ingestion/standalone-realtime.html">Realtime Process</a></li></ul></div></div></section></div><script> |
| var coll = document.getElementsByClassName('collapsible'); |
| var checkActiveCategory = true; |
| for (var i = 0; i < coll.length; i++) { |
| var links = coll[i].nextElementSibling.getElementsByTagName('*'); |
| if (checkActiveCategory){ |
| for (var j = 0; j < links.length; j++) { |
| if (links[j].classList.contains('navListItemActive')){ |
| coll[i].nextElementSibling.classList.toggle('hide'); |
| coll[i].childNodes[1].classList.toggle('rotate'); |
| checkActiveCategory = false; |
| break; |
| } |
| } |
| } |
| |
| coll[i].addEventListener('click', function() { |
| var arrow = this.childNodes[1]; |
| arrow.classList.toggle('rotate'); |
| var content = this.nextElementSibling; |
| content.classList.toggle('hide'); |
| }); |
| } |
| |
| document.addEventListener('DOMContentLoaded', function() { |
| createToggler('#navToggler', '#docsNav', 'docsSliderActive'); |
| createToggler('#tocToggler', 'body', 'tocActive'); |
| |
| var headings = document.querySelector('.toc-headings'); |
| headings && headings.addEventListener('click', function(event) { |
| var el = event.target; |
| while(el !== headings){ |
| if (el.tagName === 'A') { |
| document.body.classList.remove('tocActive'); |
| break; |
| } else{ |
| el = el.parentNode; |
| } |
| } |
| }, false); |
| |
| function createToggler(togglerSelector, targetSelector, className) { |
| var toggler = document.querySelector(togglerSelector); |
| var target = document.querySelector(targetSelector); |
| |
| if (!toggler) { |
| return; |
| } |
| |
| toggler.onclick = function(event) { |
| event.preventDefault(); |
| |
| target.classList.toggle(className); |
| }; |
| } |
| }); |
| </script></nav></div><div class="container mainContainer docsContainer"><div class="wrapper"><div class="post"><header class="postHeader"><a class="edit-page-link button" href="https://github.com/apache/druid/edit/master/docs/multi-stage-query/concepts.md" target="_blank" rel="noreferrer noopener">Edit</a><h1 id="__docusaurus" class="postHeaderTitle">SQL-based ingestion concepts</h1></header><article><div><span><!-- |
| ~ Licensed to the Apache Software Foundation (ASF) under one |
| ~ or more contributor license agreements. See the NOTICE file |
| ~ distributed with this work for additional information |
| ~ regarding copyright ownership. The ASF licenses this file |
| ~ to you under the Apache License, Version 2.0 (the |
| ~ "License"); you may not use this file except in compliance |
| ~ with the License. You may obtain a copy of the License at |
| ~ |
| ~ http://www.apache.org/licenses/LICENSE-2.0 |
| ~ |
| ~ Unless required by applicable law or agreed to in writing, |
| ~ software distributed under the License is distributed on an |
| ~ "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| ~ KIND, either express or implied. See the License for the |
| ~ specific language governing permissions and limitations |
| ~ under the License. |
| --> |
| <blockquote> |
| <p>This page describes SQL-based batch ingestion using the <a href="/docs/26.0.0/multi-stage-query/index.html"><code>druid-multi-stage-query</code></a> |
| extension, new in Druid 24.0. Refer to the <a href="/docs/26.0.0/ingestion/index.html#batch">ingestion methods</a> table to determine which |
| ingestion method is right for you.</p> |
| </blockquote> |
| <h2><a class="anchor" aria-hidden="true" id="multi-stage-query-task-engine"></a><a href="#multi-stage-query-task-engine" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Multi-stage query task engine</h2> |
| <p>The <code>druid-multi-stage-query</code> extension adds a multi-stage query (MSQ) task engine that executes SQL statements as batch |
| tasks in the indexing service, which execute on <a href="/docs/26.0.0/design/architecture.html#druid-services">Middle Managers</a>. |
| <a href="/docs/26.0.0/multi-stage-query/reference.html#insert">INSERT</a> and <a href="/docs/26.0.0/multi-stage-query/reference.html#replace">REPLACE</a> tasks publish |
| <a href="/docs/26.0.0/design/architecture.html#datasources-and-segments">segments</a> just like <a href="/docs/26.0.0/ingestion/index.html#batch">all other forms of batch |
| ingestion</a>. Each query occupies at least two task slots while running: one controller task, |
| and at least one worker task. As an experimental feature, the MSQ task engine also supports running SELECT queries as |
| batch tasks. The behavior and result format of plain SELECT (without INSERT or REPLACE) is subject to change.</p> |
| <p>You can execute SQL statements using the MSQ task engine through the <strong>Query</strong> view in the <a href="/docs/26.0.0/operations/web-console.html">web |
| console</a> or through the <a href="/docs/26.0.0/multi-stage-query/api.html"><code>/druid/v2/sql/task</code> API</a>.</p> |
| <p>For more details on how SQL queries are executed using the MSQ task engine, see <a href="#multi-stage-query-tasks">multi-stage query |
| tasks</a>.</p> |
| <h2><a class="anchor" aria-hidden="true" id="sql-extensions"></a><a href="#sql-extensions" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>SQL extensions</h2> |
| <p>To support ingestion, additional SQL functionality is available through the MSQ task engine.</p> |
| <p><a name="extern"></a></p> |
| <h3><a class="anchor" aria-hidden="true" id="read-external-data-with-extern"></a><a href="#read-external-data-with-extern" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Read external data with <code>EXTERN</code></h3> |
| <p>Query tasks can access external data through the <code>EXTERN</code> function, using any native batch <a href="/docs/26.0.0/ingestion/native-batch-input-sources.html">input |
| source</a> and <a href="/docs/26.0.0/ingestion/data-formats.html#input-format">input format</a>.</p> |
| <p><code>EXTERN</code> can read multiple files in parallel across different worker tasks. However, <code>EXTERN</code> does not split individual |
| files across multiple worker tasks. If you have a small number of very large input files, you can increase query |
| parallelism by splitting up your input files.</p> |
| <p>For more information about the syntax, see <a href="/docs/26.0.0/multi-stage-query/reference.html#extern-function"><code>EXTERN</code></a>.</p> |
| <p>See also the set of SQL-friendly input-source-specific table functions which may be more convenient |
| than <code>EXTERN</code>.</p> |
| <p><a name="insert"></a></p> |
| <h3><a class="anchor" aria-hidden="true" id="load-data-with-insert"></a><a href="#load-data-with-insert" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Load data with <code>INSERT</code></h3> |
| <p><code>INSERT</code> statements can create a new datasource or append to an existing datasource. In Druid SQL, unlike standard SQL, |
| there is no syntactical difference between creating a table and appending data to a table. Druid does not include a |
| <code>CREATE TABLE</code> statement.</p> |
| <p>Nearly all <code>SELECT</code> capabilities are available for <code>INSERT ... SELECT</code> queries. Certain exceptions are listed on the <a href="/docs/26.0.0/multi-stage-query/known-issues.html#select-statement">Known |
| issues</a> page.</p> |
| <p><code>INSERT</code> statements acquire a shared lock to the target datasource. Multiple <code>INSERT</code> statements can run at the same time, |
| for the same datasource, if your cluster has enough task slots.</p> |
| <p>Like all other forms of <a href="/docs/26.0.0/ingestion/index.html#batch">batch ingestion</a>, each <code>INSERT</code> statement generates new segments and |
| publishes them at the end of its run. For this reason, it is best suited to loading data in larger batches. Do not use |
| <code>INSERT</code> statements to load data in a sequence of microbatches; for that, use <a href="/docs/26.0.0/ingestion/index.html#streaming">streaming |
| ingestion</a> instead.</p> |
| <p>When deciding whether to use <code>REPLACE</code> or <code>INSERT</code>, keep in mind that segments generated with <code>REPLACE</code> can be pruned |
| with dimension-based pruning but those generated with <code>INSERT</code> cannot. For more information about the requirements |
| for dimension-based pruning, see <a href="#clustering">Clustering</a>.</p> |
| <p>For more information about the syntax, see <a href="/docs/26.0.0/multi-stage-query/reference.html#insert">INSERT</a>.</p> |
| <p><a name="replace"></a></p> |
| <h3><a class="anchor" aria-hidden="true" id="overwrite-data-with-replace"></a><a href="#overwrite-data-with-replace" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Overwrite data with REPLACE</h3> |
| <p><code>REPLACE</code> statements can create a new datasource or overwrite data in an existing datasource. In Druid SQL, unlike |
| standard SQL, there is no syntactical difference between creating a table and overwriting data in a table. Druid does |
| not include a <code>CREATE TABLE</code> statement.</p> |
| <p><code>REPLACE</code> uses an <a href="/docs/26.0.0/multi-stage-query/reference.html#replace-specific-time-ranges">OVERWRITE clause</a> to determine which data to overwrite. You |
| can overwrite an entire table, or a specific time range of a table. When you overwrite a specific time range, that time |
| range must align with the granularity specified in the <code>PARTITIONED BY</code> clause.</p> |
| <p><code>REPLACE</code> statements acquire an exclusive write lock to the target time range of the target datasource. No other ingestion |
| or compaction operations may proceed for that time range while the task is running. However, ingestion and compaction |
| operations may proceed for other time ranges.</p> |
| <p>Nearly all <code>SELECT</code> capabilities are available for <code>REPLACE ... SELECT</code> queries. Certain exceptions are listed on the <a href="/docs/26.0.0/multi-stage-query/known-issues.html#select-statement">Known |
| issues</a> page.</p> |
| <p>For more information about the syntax, see <a href="/docs/26.0.0/multi-stage-query/reference.html#replace">REPLACE</a>.</p> |
| <p>When deciding whether to use <code>REPLACE</code> or <code>INSERT</code>, keep in mind that segments generated with <code>REPLACE</code> can be pruned |
| with dimension-based pruning but those generated with <code>INSERT</code> cannot. For more information about the requirements |
| for dimension-based pruning, see <a href="#clustering">Clustering</a>.</p> |
| <h3><a class="anchor" aria-hidden="true" id="primary-timestamp"></a><a href="#primary-timestamp" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Primary timestamp</h3> |
| <p>Druid tables always include a primary timestamp named <code>__time</code>.</p> |
| <p>It is common to set a primary timestamp by using <a href="/docs/26.0.0/querying/sql-scalar.html#date-and-time-functions">date and time |
| functions</a>; for example: <code>TIME_FORMAT("timestamp", 'yyyy-MM-dd HH:mm:ss') AS __time</code>.</p> |
| <p>The <code>__time</code> column is used for <a href="#partitioning-by-time">partitioning by time</a>. If you use <code>PARTITIONED BY ALL</code> or |
| <code>PARTITIONED BY ALL TIME</code>, partitioning by time is disabled. In these cases, you do not need to include a <code>__time</code> |
| column in your <code>INSERT</code> statement. However, Druid still creates a <code>__time</code> column in your Druid table and sets all |
| timestamps to 1970-01-01 00:00:00.</p> |
| <p>For more information, see <a href="/docs/26.0.0/ingestion/data-model.html#primary-timestamp">Primary timestamp</a>.</p> |
| <p><a name="partitioning"></a></p> |
| <h3><a class="anchor" aria-hidden="true" id="partitioning-by-time"></a><a href="#partitioning-by-time" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Partitioning by time</h3> |
| <p><code>INSERT</code> and <code>REPLACE</code> statements require the <code>PARTITIONED BY</code> clause, which determines how time-based partitioning is done. |
| In Druid, data is split into one or more segments per time chunk, defined by the PARTITIONED BY granularity.</p> |
| <p>Partitioning by time is important for three reasons:</p> |
| <ol> |
| <li>Queries that filter by <code>__time</code> (SQL) or <code>intervals</code> (native) are able to use time partitioning to prune the set of |
| segments to consider.</li> |
| <li>Certain data management operations, such as overwriting and compacting existing data, acquire exclusive write locks |
| on time partitions. Finer-grained partitioning allows finer-grained exclusive write locks.</li> |
| <li>Each segment file is wholly contained within a time partition. Too-fine-grained partitioning may cause a large number |
| of small segments, which leads to poor performance.</li> |
| </ol> |
| <p><code>PARTITIONED BY HOUR</code> and <code>PARTITIONED BY DAY</code> are the most common choices to balance these considerations. <code>PARTITIONED BY ALL</code> is suitable if your dataset does not have a <a href="#primary-timestamp">primary timestamp</a>.</p> |
| <p>For more information about the syntax, see <a href="/docs/26.0.0/multi-stage-query/reference.html#partitioned-by">PARTITIONED BY</a>.</p> |
| <h3><a class="anchor" aria-hidden="true" id="clustering"></a><a href="#clustering" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Clustering</h3> |
| <p>Within each time chunk defined by <a href="#partitioning-by-time">time partitioning</a>, data can be further split by the optional |
| <a href="/docs/26.0.0/multi-stage-query/reference.html#clustered-by">CLUSTERED BY</a> clause.</p> |
| <p>For example, suppose you ingest 100 million rows per hour using <code>PARTITIONED BY HOUR</code> and <code>CLUSTERED BY hostName</code>. The |
| ingestion task will generate segments of roughly 3 million rows โ the default value of |
| <a href="/docs/26.0.0/multi-stage-query/reference.html#context-parameters"><code>rowsPerSegment</code></a> โ with lexicographic ranges of <code>hostName</code>s grouped into segments.</p> |
| <p>Clustering is important for two reasons:</p> |
| <ol> |
| <li>Lower storage footprint due to improved locality, and therefore improved compressibility.</li> |
| <li>Better query performance due to dimension-based segment pruning, which removes segments from consideration when they |
| cannot possibly contain data matching a query's filter. This speeds up filters like <code>x = 'foo'</code> and <code>x IN ('foo', 'bar')</code>.</li> |
| </ol> |
| <p>To activate dimension-based pruning, these requirements must be met:</p> |
| <ul> |
| <li>Segments were generated by a <code>REPLACE</code> statement, not an <code>INSERT</code> statement.</li> |
| <li>All <code>CLUSTERED BY</code> columns are single-valued string columns.</li> |
| </ul> |
| <p>If these requirements are <em>not</em> met, Druid still clusters data during ingestion but will not be able to perform |
| dimension-based segment pruning at query time. You can tell if dimension-based segment pruning is possible by using the |
| <code>sys.segments</code> table to inspect the <code>shard_spec</code> for the segments generated by an ingestion query. If they are of type |
| <code>range</code> or <code>single</code>, then dimension-based segment pruning is possible. Otherwise, it is not. The shard spec type is also |
| available in the <strong>Segments</strong> view under the <strong>Partitioning</strong> column.</p> |
| <p>For more information about syntax, see <a href="/docs/26.0.0/multi-stage-query/reference.html#clustered-by"><code>CLUSTERED BY</code></a>.</p> |
| <h3><a class="anchor" aria-hidden="true" id="rollup"></a><a href="#rollup" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Rollup</h3> |
| <p><a href="/docs/26.0.0/ingestion/rollup.html">Rollup</a> is a technique that pre-aggregates data during ingestion to reduce the amount of data |
| stored. Intermediate aggregations are stored in the generated segments, and further aggregation is done at query time. |
| This reduces storage footprint and improves performance, often dramatically.</p> |
| <p>To perform ingestion with rollup:</p> |
| <ol> |
| <li>Use <code>GROUP BY</code>. The columns in the <code>GROUP BY</code> clause become dimensions, and aggregation functions become metrics.</li> |
| <li>Set <a href="/docs/26.0.0/multi-stage-query/reference.html#context-parameters"><code>finalizeAggregations: false</code></a> in your context. This causes aggregation |
| functions to write their internal state to the generated segments, instead of the finalized end result, and enables |
| further aggregation at query time.</li> |
| <li>Wrap all multi-value strings in <code>MV_TO_ARRAY(...)</code> and set <a href="/docs/26.0.0/multi-stage-query/reference.html#context-parameters"><code>groupByEnableMultiValueUnnesting: false</code></a> in your context. This ensures that multi-value strings are left alone and |
| remain lists, instead of being <a href="/docs/26.0.0/querying/sql-data-types.html#multi-value-strings">automatically unnested</a> by the |
| <code>GROUP BY</code> operator.</li> |
| </ol> |
| <p>When you do all of these things, Druid understands that you intend to do an ingestion with rollup, and it writes |
| rollup-related metadata into the generated segments. Other applications can then use <a href="/docs/26.0.0/querying/segmentmetadataquery.html"><code>segmentMetadata</code> |
| queries</a> to retrieve rollup-related information.</p> |
| <p>If you see the error "Encountered multi-value dimension <code>x</code> that cannot be processed with |
| groupByEnableMultiValueUnnesting set to false", then wrap that column in <code>MV_TO_ARRAY(x) AS x</code>.</p> |
| <p>The following <a href="/docs/26.0.0/querying/sql-aggregations.html">aggregation functions</a> are supported for rollup at ingestion time: |
| <code>COUNT</code> (but switch to <code>SUM</code> at query time), <code>SUM</code>, <code>MIN</code>, <code>MAX</code>, <code>EARLIEST</code> (<a href="/docs/26.0.0/multi-stage-query/known-issues.html#select-statement">string only</a>), |
| <code>LATEST</code> (<a href="/docs/26.0.0/multi-stage-query/known-issues.html#select-statement">string only</a>), <code>APPROX_COUNT_DISTINCT</code>, <code>APPROX_COUNT_DISTINCT_BUILTIN</code>, |
| <code>APPROX_COUNT_DISTINCT_DS_HLL</code>, <code>APPROX_COUNT_DISTINCT_DS_THETA</code>, and <code>DS_QUANTILES_SKETCH</code> (but switch to |
| <code>APPROX_QUANTILE_DS</code> at query time). Do not use <code>AVG</code>; instead, use <code>SUM</code> and <code>COUNT</code> at ingest time and compute the |
| quotient at query time.</p> |
| <p>For an example, see <a href="/docs/26.0.0/multi-stage-query/examples.html#insert-with-rollup">INSERT with rollup example</a>.</p> |
| <h2><a class="anchor" aria-hidden="true" id="multi-stage-query-tasks"></a><a href="#multi-stage-query-tasks" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Multi-stage query tasks</h2> |
| <h3><a class="anchor" aria-hidden="true" id="execution-flow"></a><a href="#execution-flow" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Execution flow</h3> |
| <p>When you execute a SQL statement using the task endpoint <a href="/docs/26.0.0/multi-stage-query/api.html#submit-a-query"><code>/druid/v2/sql/task</code></a>, the following |
| happens:</p> |
| <ol> |
| <li><p>The Broker plans your SQL query into a native query, as usual.</p></li> |
| <li><p>The Broker wraps the native query into a task of type <code>query_controller</code> |
| and submits it to the indexing service.</p></li> |
| <li><p>The Broker returns the task ID to you and exits.</p></li> |
| <li><p>The controller task launches some number of worker tasks determined by |
| the <code>maxNumTasks</code> and <code>taskAssignment</code> <a href="/docs/26.0.0/multi-stage-query/reference.html#context-parameters">context parameters</a>. You can set these settings individually for each query.</p></li> |
| <li><p>Worker tasks of type <code>query_worker</code> execute the query.</p></li> |
| <li><p>If the query is a <code>SELECT</code> query, the worker tasks send the results |
| back to the controller task, which writes them into its task report. |
| If the query is an INSERT or REPLACE query, the worker tasks generate and |
| publish new Druid segments to the provided datasource.</p></li> |
| </ol> |
| <h3><a class="anchor" aria-hidden="true" id="parallelism"></a><a href="#parallelism" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Parallelism</h3> |
| <p>The <a href="/docs/26.0.0/multi-stage-query/reference.html#context-parameters"><code>maxNumTasks</code></a> query parameter determines the maximum number of tasks your |
| query will use, including the one <code>query_controller</code> task. Generally, queries perform better with more workers. The |
| lowest possible value of <code>maxNumTasks</code> is two (one worker and one controller). Do not set this higher than the number of |
| free slots available in your cluster; doing so will result in a <a href="/docs/26.0.0/multi-stage-query/reference.html#error_TaskStartTimeout">TaskStartTimeout</a> |
| error.</p> |
| <p>When <a href="#extern">reading external data</a>, EXTERN can read multiple files in parallel across |
| different worker tasks. However, EXTERN does not split individual files across multiple worker tasks. If you have a |
| small number of very large input files, you can increase query parallelism by splitting up your input files.</p> |
| <p>The <code>druid.worker.capacity</code> server property on each <a href="/docs/26.0.0/design/architecture.html#druid-services">Middle Manager</a> |
| determines the maximum number of worker tasks that can run on each server at once. Worker tasks run single-threaded, |
| which also determines the maximum number of processors on the server that can contribute towards multi-stage queries.</p> |
| <h3><a class="anchor" aria-hidden="true" id="memory-usage"></a><a href="#memory-usage" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Memory usage</h3> |
| <p>Increasing the amount of available memory can improve performance in certain cases:</p> |
| <ul> |
| <li>Segment generation becomes more efficient when data doesn't spill to disk as often.</li> |
| <li>Sorting stage output data becomes more efficient since available memory affects the |
| number of required sorting passes.</li> |
| </ul> |
| <p>Worker tasks use both JVM heap memory and off-heap ("direct") memory.</p> |
| <p>On Peons launched by Middle Managers, the bulk of the JVM heap (75%, less any space used by |
| <a href="/docs/26.0.0/querying/lookups.html">lookups</a>) is split up into two bundles of equal size: one processor bundle and one worker |
| bundle. Each one comprises 37.5% of the available JVM heap, less any space used by <a href="/docs/26.0.0/querying/lookups.html">lookups</a>.</p> |
| <p>Depending on the type of query, controller and worker tasks may use sketches for determining partition boundaries. |
| The heap footprint of these sketches is capped at 10% of available memory, or 300 MB, whichever is lower.</p> |
| <p>The processor memory bundle is used for query processing and segment generation. Each processor bundle must also |
| provides space to buffer I/O between stages. Specifically, each downstream stage requires 1 MB of buffer space for each |
| upstream worker. For example, if you have 100 workers running in stage 0, and stage 1 reads from stage 0, then each |
| worker in stage 1 requires 1M * 100 = 100 MB of memory for frame buffers.</p> |
| <p>The worker memory bundle is used for sorting stage output data prior to shuffle. Workers can sort more data than fits in |
| memory; in this case, they will switch to using disk.</p> |
| <p>Worker tasks also use off-heap ("direct") memory. Set the amount of direct memory available (<code>-XX:MaxDirectMemorySize</code>) |
| to at least <code>(druid.processing.numThreads + 1) * druid.processing.buffer.sizeBytes</code>. Increasing the amount of direct |
| memory available beyond the minimum does not speed up processing.</p> |
| <h3><a class="anchor" aria-hidden="true" id="disk-usage"></a><a href="#disk-usage" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>Disk usage</h3> |
| <p>Worker tasks use local disk for four purposes:</p> |
| <ul> |
| <li>Temporary copies of input data. Each temporary file is deleted before the next one is read. You only need |
| enough temporary disk space to store one input file at a time per task.</li> |
| <li>Temporary data related to segment generation. You only need enough temporary disk space to store one segments' worth |
| of data at a time per task. This is generally less than 2 GB per task.</li> |
| <li>External sort of data prior to shuffle. Requires enough space to store a compressed copy of the entire output dataset |
| for a task.</li> |
| <li>Storing stage output data during a shuffle. Requires enough space to store a compressed copy of the entire output |
| dataset for a task.</li> |
| </ul> |
| <p>Workers use the task working directory, given by |
| <a href="/docs/26.0.0/configuration/index.html#additional-peon-configuration"><code>druid.indexer.task.baseDir</code></a>, for these items. It is |
| important that this directory has enough space available for these purposes.</p> |
| </span></div></article></div><div class="docs-prevnext"><a class="docs-prev button" href="/docs/26.0.0/multi-stage-query/index.html"><span class="arrow-prev">โ </span><span>Overview</span></a><a class="docs-next button" href="/docs/26.0.0/multi-stage-query/api.html"><span>API</span><span class="arrow-next"> โ</span></a></div></div></div><nav class="onPageNav"><ul class="toc-headings"><li><a href="#multi-stage-query-task-engine">Multi-stage query task engine</a></li><li><a href="#sql-extensions">SQL extensions</a><ul class="toc-headings"><li><a href="#read-external-data-with-extern">Read external data with <code>EXTERN</code></a></li><li><a href="#load-data-with-insert">Load data with <code>INSERT</code></a></li><li><a href="#overwrite-data-with-replace">Overwrite data with REPLACE</a></li><li><a href="#primary-timestamp">Primary timestamp</a></li><li><a href="#partitioning-by-time">Partitioning by time</a></li><li><a href="#clustering">Clustering</a></li><li><a href="#rollup">Rollup</a></li></ul></li><li><a href="#multi-stage-query-tasks">Multi-stage query tasks</a><ul class="toc-headings"><li><a href="#execution-flow">Execution flow</a></li><li><a href="#parallelism">Parallelism</a></li><li><a href="#memory-usage">Memory usage</a></li><li><a href="#disk-usage">Disk usage</a></li></ul></li></ul></nav></div><footer class="nav-footer druid-footer" id="footer"><div class="container"><div class="text-center"><p><a href="/technology">Technology</a>โยทโ<a href="/use-cases">Use Cases</a>โยทโ<a href="/druid-powered">Powered by Druid</a>โยทโ<a href="/docs/26.0.0/">Docs</a>โยทโ<a href="/community/">Community</a>โยทโ<a href="/downloads.html">Download</a>โยทโ<a href="/faq">FAQ</a></p></div><div class="text-center"><a title="Join the user group" href="https://groups.google.com/forum/#!forum/druid-user" target="_blank"><span class="fa fa-comments"></span></a>โยทโ<a title="Follow Druid" href="https://twitter.com/druidio" target="_blank"><span class="fab fa-twitter"></span></a>โยทโ<a title="Download via Apache" href="https://www.apache.org/dyn/closer.cgi?path=/incubator/druid/{{ site.druid_versions[0].versions[0].version }}/apache-druid-{{ site.druid_versions[0].versions[0].version }}-bin.tar.gz" target="_blank"><span class="fas fa-feather"></span></a>โยทโ<a title="GitHub" href="https://github.com/apache/druid" target="_blank"><span class="fab fa-github"></span></a></div><div class="text-center license">Copyright ยฉ 2022 <a href="https://www.apache.org/" target="_blank">Apache Software Foundation</a>.<br/>Except where otherwise noted, licensed under <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA 4.0</a>.<br/>Apache Druid, Druid, and the Druid logo are either registered trademarks or trademarks of The Apache Software Foundation in the United States and other countries.</div></div></footer></div><script type="text/javascript" src="https://cdn.jsdelivr.net/npm/docsearch.js@2/dist/cdn/docsearch.min.js"></script><script> |
| document.addEventListener('keyup', function(e) { |
| if (e.target !== document.body) { |
| return; |
| } |
| // keyCode for '/' (slash) |
| if (e.keyCode === 191) { |
| const search = document.getElementById('search_input_react'); |
| search && search.focus(); |
| } |
| }); |
| </script><script> |
| var search = docsearch({ |
| appId: 'CPK9PMSCEY', |
| apiKey: 'd4ef4ffe3a2f0c7d1e34b062fd98736b', |
| indexName: 'apache_druid', |
| inputSelector: '#search_input_react', |
| algoliaOptions: {"facetFilters":["language:en","version:26.0.0"]} |
| }); |
| </script></body></html> |