[
https://issues.apache.org/jira/browse/DRILL-5815?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16195481#comment-16195481
]
ASF GitHub Bot commented on DRILL-5815:
---------------------------------------
Github user Ben-Zvi commented on a diff in the pull request:
https://github.com/apache/drill/pull/960#discussion_r143317064
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/util/MemoryAllocationUtilities.java
---
@@ -56,30 +59,90 @@ public static void
setupBufferedOpsMemoryAllocations(final PhysicalPlan plan, fi
}
// if there are any sorts, compute the maximum allocation, and set it
on them
- if (bufferedOpList.size() > 0) {
- final OptionManager optionManager = queryContext.getOptions();
- double cpu_load_average =
optionManager.getOption(ExecConstants.CPU_LOAD_AVERAGE);
- final long maxWidth =
optionManager.getOption(ExecConstants.MAX_WIDTH_PER_NODE);
- final long maxWidthPerNode =
ExecConstants.MAX_WIDTH_PER_NODE.computeMaxWidth(cpu_load_average,maxWidth);
- long maxAllocPerNode = Math.min(DrillConfig.getMaxDirectMemory(),
-
queryContext.getConfig().getLong(RootAllocatorFactory.TOP_LEVEL_MAX_ALLOC));
- maxAllocPerNode = Math.min(maxAllocPerNode,
-
optionManager.getOption(ExecConstants.MAX_QUERY_MEMORY_PER_NODE_KEY).num_val);
- final long maxOperatorAlloc = maxAllocPerNode /
(bufferedOpList.size() * maxWidthPerNode);
- logger.debug("Max buffered operator alloc: {}", maxOperatorAlloc);
-
- // User configurable option to allow forcing minimum memory.
- // Ensure that the buffered ops receive the minimum memory needed to
make progress.
- // Without this, the math might work out to allocate too little
memory.
- final long opMinMem =
queryContext.getOptions().getOption(ExecConstants.MIN_MEMORY_PER_BUFFERED_OP_KEY).num_val;
-
- for(final PhysicalOperator op : bufferedOpList) {
-
- long alloc = Math.max(maxOperatorAlloc, op.getInitialAllocation());
- alloc = Math.max(alloc, opMinMem);
- op.setMaxAllocation(alloc);
- }
- }
plan.getProperties().hasResourcePlan = true;
+ if (bufferedOpList.isEmpty()) {
+ return;
+ }
+
+ // Setup options, etc.
+
+ final OptionManager optionManager = queryContext.getOptions();
+ final long directMemory = DrillConfig.getMaxDirectMemory();
+
+ // Compute per-node, per-query memory.
+
+ final long maxAllocPerNode =
computeQueryMemory(queryContext.getConfig(), optionManager, directMemory);
+ logger.debug("Memory per query per node: {}", maxAllocPerNode);
+
+ // Now divide up the memory by slices and operators.
+
+ final long opMinMem = computeOperatorMemory(optionManager,
maxAllocPerNode, bufferedOpList.size());
--- End diff --
This seems like a short term idea (MEP 4 ?) ; else later the Hash Join
would become a "buffered operator", and possibly we'd use a dynamic memory
scheme (i.e., allocate minimum to each operator, and add more to those in need)
which would cover this issue better.
> Provide option to set query memory as percent of total
> ------------------------------------------------------
>
> Key: DRILL-5815
> URL: https://issues.apache.org/jira/browse/DRILL-5815
> Project: Apache Drill
> Issue Type: Improvement
> Affects Versions: 1.11.0
> Reporter: Paul Rogers
> Assignee: Paul Rogers
> Fix For: 1.12.0
>
>
> Drill provides a parameter to set the memory per query as a static number
> which defaults to 2 GB. This number is a wonderful setting for the default
> Drillbit configuration of 8 GB heap; it allows 2-3 concurrent queries. But,
> as Drillbit memory increases, the default becomes a bit constraining. While
> users can change the setting, they seldom do.
> In addition, provide an option that sets memory as a percent of total memory.
> If the allocation is 10%, say, and total memory is 128 GB, then each query
> gets ~13GB, which is a big improvement.
> The existing option acts as a floor: the query must receive at least that
> much memory.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
