<?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="mem_limit">

  <title>MEM_LIMIT Query Option</title>

  <titlealts audience="PDF">

    <navtitle>MEM_LIMIT</navtitle>

  </titlealts>

  <prolog>
    <metadata>
      <data name="Category" value="Impala"/>
      <data name="Category" value="Impala Query Options"/>
      <data name="Category" value="Scalability"/>
      <data name="Category" value="Memory"/>
      <data name="Category" value="Troubleshooting"/>
      <data name="Category" value="Developers"/>
      <data name="Category" value="Data Analysts"/>
    </metadata>
  </prolog>

  <conbody>

    <p>
      The MEM_LIMIT query option defines the maximum amount of memory a query can allocate on
      each node. The total memory that can be used by a query is the <codeph>MEM_LIMIT</codeph>
      times the number of nodes.
    </p>

    <p rev="">
      There are two levels of memory limit for Impala. The
      <codeph>&#8209;&#8209;mem_limit</codeph> startup option sets an overall limit for the
      <cmdname>impalad</cmdname> process (which handles multiple queries concurrently). That
      process memory limit can be expressed either as a percentage of RAM available to the
      process such as <codeph>&#8209;&#8209;mem_limit=70%</codeph> or as a fixed amount of
      memory, such as <codeph>100gb</codeph>. The memory available to the process is based on
      the host's physical memory and, since Impala 3.2, memory limits from Linux Control Groups.
      E.g. if an <cmdname>impalad</cmdname> process is running in a Docker container on a host
      with 100GB of memory, the memory available is 100GB or the Docker container's memory
      limit, whichever is less.
    </p>

    <p rev="">
      The <codeph>MEM_LIMIT</codeph> query option, which you set through
      <cmdname>impala-shell</cmdname> or the <codeph>SET</codeph> statement in a JDBC or ODBC
      application, applies to each individual query. The <codeph>MEM_LIMIT</codeph> query option
      is usually expressed as a fixed size such as <codeph>10gb</codeph>, and must always be
      less than the <cmdname>impalad</cmdname> memory limit.
    </p>

    <p rev="">
      If query processing exceeds the specified memory limit on any node, either the per-query
      limit or the <cmdname>impalad</cmdname> limit, Impala cancels the query automatically.
      Memory limits are checked periodically during query processing, so the actual memory in
      use might briefly exceed the limit without the query being cancelled.
    </p>

    <p>
      <b>Type:</b> numeric
    </p>

    <p rev="">
      <b>Units:</b> A numeric argument represents memory size in bytes; you can also use a
      suffix of <codeph>m</codeph> or <codeph>mb</codeph> for megabytes, or more commonly
      <codeph>g</codeph> or <codeph>gb</codeph> for gigabytes. If you specify a value with
      unrecognized formats, subsequent queries fail with an error.
    </p>

    <p rev="">
      <b>Default:</b> 0 (unlimited)
    </p>

    <p conref="../shared/impala_common.xml#common/usage_notes_blurb"/>

    <p rev="">
      The <codeph>MEM_LIMIT</codeph> setting is primarily useful in a high-concurrency setting,
      or on a cluster with a workload shared between Impala and other data processing
      components. You can prevent any query from accidentally using much more memory than
      expected, which could negatively impact other Impala queries.
    </p>

    <p rev="">
      Use the output of the <codeph>SUMMARY</codeph> command in <cmdname>impala-shell</cmdname>
      to get a report of memory used for each phase of your most heavyweight queries on each
      node, and then set a <codeph>MEM_LIMIT</codeph> somewhat higher than that. See
      <xref href="impala_explain_plan.xml#perf_summary"/> for usage information about the
      <codeph>SUMMARY</codeph> command.
    </p>

    <p conref="../shared/impala_common.xml#common/example_blurb" rev=""/>

    <p rev="">
      The following examples show how to set the <codeph>MEM_LIMIT</codeph> query option using a
      fixed number of bytes, or suffixes representing gigabytes or megabytes.
    </p>

<codeblock rev="">
[localhost:21000] > set mem_limit=3000000000;
MEM_LIMIT set to 3000000000
[localhost:21000] > select 5;
Query: select 5
+---+
| 5 |
+---+
| 5 |
+---+

[localhost:21000] > set mem_limit=3g;
MEM_LIMIT set to 3g
[localhost:21000] > select 5;
Query: select 5
+---+
| 5 |
+---+
| 5 |
+---+

[localhost:21000] > set mem_limit=3gb;
MEM_LIMIT set to 3gb
[localhost:21000] > select 5;
+---+
| 5 |
+---+
| 5 |
+---+

[localhost:21000] > set mem_limit=3m;
MEM_LIMIT set to 3m
[localhost:21000] > select 5;
+---+
| 5 |
+---+
| 5 |
+---+
[localhost:21000] > set mem_limit=3mb;
MEM_LIMIT set to 3mb
[localhost:21000] > select 5;
+---+
| 5 |
+---+
</codeblock>

    <p rev="">
      The following examples show how unrecognized <codeph>MEM_LIMIT</codeph> values lead to
      errors for subsequent queries.
    </p>

<codeblock rev="">
[localhost:21000] > set mem_limit=3pb;
MEM_LIMIT set to 3pb
[localhost:21000] > select 5;
ERROR: Failed to parse query memory limit from '3pb'.

[localhost:21000] > set mem_limit=xyz;
MEM_LIMIT set to xyz
[localhost:21000] > select 5;
Query: select 5
ERROR: Failed to parse query memory limit from 'xyz'.
</codeblock>

    <p rev="">
      The following examples shows the automatic query cancellation when the
      <codeph>MEM_LIMIT</codeph> value is exceeded on any host involved in the Impala query.
      First it runs a successful query and checks the largest amount of memory used on any node
      for any stage of the query. Then it sets an artificially low <codeph>MEM_LIMIT</codeph>
      setting so that the same query cannot run.
    </p>

<codeblock rev="">
[localhost:21000] > select count(*) from customer;
Query: select count(*) from customer
+----------+
| count(*) |
+----------+
| 150000   |
+----------+

[localhost:21000] > select count(distinct c_name) from customer;
Query: select count(distinct c_name) from customer
+------------------------+
| count(distinct c_name) |
+------------------------+
| 150000                 |
+------------------------+

[localhost:21000] > summary;
+--------------+--------+--------+----------+----------+---------+------------+----------+---------------+---------------+
| Operator     | #Hosts | #Inst  | Avg Time | Max Time | #Rows   | Est. #Rows | Peak Mem | Est. Peak Mem | Detail        |
+--------------+--------+--------+----------+----------+---------+------------+----------+---------------+---------------+
| 06:AGGREGATE | 1      | 1      | 230.00ms | 230.00ms | 1       | 1          | 16.00 KB | -1 B          | FINALIZE      |
| 05:EXCHANGE  | 1      | 1      | 43.44us  | 43.44us  | 1       | 1          | 0 B      | -1 B          | UNPARTITIONED |
| 02:AGGREGATE | 1      | 1      | 227.14ms | 227.14ms | 1       | 1          | 12.00 KB | 10.00 MB      |               |
| 04:AGGREGATE | 1      | 1      | 126.27ms | 126.27ms | 150.00K | 150.00K    | 15.17 MB | 10.00 MB      |               |
| 03:EXCHANGE  | 1      | 1      | 44.07ms  | 44.07ms  | 150.00K | 150.00K    | 0 B      | 0 B           | HASH(c_name)  |
<b>| 01:AGGREGATE | 1      | 1      | 361.94ms | 361.94ms | 150.00K | 150.00K    | 23.04 MB | 10.00 MB      |               |</b>
| 00:SCAN HDFS | 1      | 1      | 43.64ms  | 43.64ms  | 150.00K | 150.00K    | 24.19 MB | 64.00 MB      | tpch.customer |
+--------------+--------+--------+----------+----------+---------+------------+----------+---------------+---------------+

[localhost:21000] > set mem_limit=15mb;
MEM_LIMIT set to 15mb
[localhost:21000] > select count(distinct c_name) from customer;
Query: select count(distinct c_name) from customer
ERROR:
Memory limit exceeded
Query did not have enough memory to get the minimum required buffers in the block manager.
</codeblock>

  </conbody>

</concept>