docs/topics/impala_scaling_limits.xml

<?xml version="1.0" encoding="UTF-8"?>
<!--
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.
-->
<!DOCTYPE concept PUBLIC "-//OASIS//DTD DITA Concept//EN" "concept.dtd">
<concept id="impala_scaling_limits">

  <title>Scaling Limits and Guidelines</title>

  <prolog>
    <metadata>
      <data name="Category" value="Impala"/>
      <data name="Category" value="Scalability"/>
    </metadata>
  </prolog>

  <conbody>

    <p>
      This topic lists the <i>scalability</i> limitation in Impala. For a given functional
      feature, it is recommended that you respect these limitations to achieve optimal
      scalability and performance. For example, you might be able to create a table with 2000
      columns, you will experience performance problems while querying the table. This topic
      does not cover functional limitations in Impala.
    </p>

    <p>
      Unless noted otherwise, the limits were tested and certified.
    </p>

    <p>
      The limits noted as "<i>generally safe</i>" are not certified, but recommended as
      generally safe. A safe range is not a hard limit as unforeseen errors or troubles in your
      particular environment can affect the range.
    </p>

    <p outputclass="toc inpage"/>

  </conbody>

  <concept id="deployment_limits">

    <title>Deployment Limits</title>

    <conbody>

      <ul>
        <li>
          Number of Impalad Executors
          <ul>
            <li>
              80 nodes in <keyword keyref="impala211"/> and lower
            </li>

            <li>
              150 nodes in <keyword keyref="impala212"/> and higher
            </li>
          </ul>
        </li>

        <li>
          Number of Impalad Coordinators: 1 coordinator for at most every 50 executors
        </li>

        <li audience="hidden">
          Max memory
        </li>

        <li audience="hidden">
          Max number of CPU cores
        </li>

        <li audience="hidden">
          Max number of disks
        </li>
      </ul>

      <ul>
        <li>
          The number of Impala clusters per deployment
          <ul>
            <li>
              1 Impala cluster in Impala 3.1 and lower
            </li>

            <li>
              Multiple clusters in Impala 3.2 and higher is <i>generally safe</i>.
            </li>
          </ul>
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="data_storage_limits">

    <title>Data Storage Limits</title>

    <conbody>

      <p>
        There are no hard limits for the following, but you will experience gradual performance
        degradation as you increase these numbers.
      </p>

      <ul>
        <li>
          Number of databases
        </li>

        <li>
          Number of tables - total, per database
        </li>

        <li>
          Number of partitions - total, per table
        </li>

        <li>
          Number of files - total, per table, per table per partition
        </li>

        <li>
          Number of views - total, per database
        </li>

        <li>
          Number of user-defined functions - total, per database
        </li>

        <li>
          Parquet
          <ul>
            <li>
              Number of columns per row group
            </li>

            <li>
              Number of row groups per block
            </li>

            <li>
              Number of HDFS blocks per file
            </li>
          </ul>
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="schema_design_limits">

    <title>Schema Design Limits</title>

    <conbody>

      <ul>
        <li>
          Number of columns
          <ul>
            <li>
              300 for Kudu tables
              <p>
                See
                <xref
                  href="https://kudu.apache.org/docs/known_issues.html"
                  format="html" scope="external">Kudu
                Usage Limitations</xref> for more information.
              </p>
            </li>

            <li>
              1000 for other types of tables
            </li>
          </ul>
        </li>

        <li audience="hidden">
          Table and column name length
        </li>

        <li audience="hidden">
          Maximum cell size
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="security_limits">

    <title>Security Limits</title>

    <conbody>

      <ul>
        <li>
          Number of roles: 10,000 for Sentry
        </li>

        <li audience="hidden">
          Number of columns used in column level ACL
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="ddl_limits" audience="hidden">

    <title>Ingestion and DDL Limits</title>

    <conbody>

      <ul>
        <li>
          Number of DDL operations per minutes
        </li>

        <li>
          Number of concurrent DDL operations
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="query_compile_limits">

    <title>Query Limits - Compile Time</title>

    <conbody>

      <ul>
        <li>
          Maximum number of columns in a query, included in a <codeph>SELECT</codeph> list,
          <codeph>INSERT</codeph>, and in an expression: no limit
        </li>

        <li>
          Number of tables referenced: no limit
        </li>

        <li>
          Number of plan nodes: no limit
        </li>

        <li>
          Number of plan fragments: no limit
        </li>

        <li>
          Depth of expression tree: 1000 hard limit
        </li>

        <li>
          Width of expression tree: 10,000 hard limit
        </li>
      </ul>

    </conbody>

  </concept>

  <concept id="query_runtime_limits">

    <title>Query Limits - Runtime Time</title>

    <conbody>

      <ul>
        <li audience="hidden">
          Number of Fragment and fragment instances
        </li>

        <li>
          Codegen
          <ul>
            <li>
              Very deeply nested expressions within queries can exceed internal Impala limits,
              leading to excessive memory usage. Setting the query option
              <codeph>disable_codegen=true</codeph> may reduce the impact, at a cost of longer
              query runtime.
            </li>
          </ul>
        </li>

        <li audience="hidden">
          Runtime Filter
          <ul>
            <li>
              Max #filter
            </li>

            <li>
              Max filter size
            </li>
          </ul>
        </li>

        <li audience="hidden">
          Query Operators
          <ul>
            <li>
              Scan
            </li>

            <li>
              Join
            </li>

            <li>
              Exchange
            </li>

            <li>
              Agg
            </li>

            <li>
              Sort
            </li>

            <li>
              Merge
            </li>
          </ul>
        </li>
      </ul>

    </conbody>

  </concept>

</concept>