This is an automated email from the ASF dual-hosted git repository.
joemcdonnell pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/impala.git
The following commit(s) were added to refs/heads/master by this push:
new 81a41fca6 IMPALA-12183: Fix cardinality clamping across aggregation
phases
81a41fca6 is described below
commit 81a41fca6dee014f11695f553ecdb6c82a172d0a
Author: Riza Suminto <[email protected]>
AuthorDate: Fri Jun 2 08:14:15 2023 -0700
IMPALA-12183: Fix cardinality clamping across aggregation phases
In the Impala planner, an aggregation node's cardinality is a sum of all
its aggregation class cardinality. An aggregation class cardinality is a
simple multiplication of NDVs of contributing grouping columns. Since
this simple multiplication of NDVs can be greater than the aggregation
node's input cardinality, each aggregation class cardinality is further
clamped at the aggregation node's input cardinality.
An aggregation operator can translate into a chain of multi-phase
aggregation plan nodes. The longest possible aggregation phase is as
follows, from the bottom to the top:
1. FIRST
2. FIRST_MERGE
3. SECOND
4. SECOND_MERGE
5. TRANSPOSE
FIRST_MERGE aggregation maintains its aggregation class cardinality
clamping at its corresponding FIRST aggregation's input
cardinality (similar relationship between SECOND_MERGE and SECOND).
However, the SECOND aggregation was clamped at the FIRST_MERGE output
cardinality instead of the FIRST input cardinality. This cardinality
mispropagation can causes cardinality explosion in the later aggregation
phase and node operator above them.
This patch fix the clamping of multi-phase aggregation to always look at
input cardinality of FIRST aggregation node. An exception is made for
TRANSPOSE phase of grouping set aggregation (such as ROLLUP). In that
case, cardinality clamping will use output cardinality of child node
right below it (either FIRST_MERGE or SECOND_MERGE) because the output
cardinality of the whole aggregation chain can be higher than input
cardinality of the FIRST phase.
Testing:
- Add test in card-agg.test
- Pass core tests.
Change-Id: I1d414fe56b027f887c7f901d8a6799a388b16b95
Reviewed-on: http://gerrit.cloudera.org:8080/20009
Reviewed-by: Aman Sinha <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
---
.../org/apache/impala/planner/AggregationNode.java | 119 ++++++++++++++++-----
.../queries/PlannerTest/card-agg.test | 100 +++++++++++++++++
.../PlannerTest/partition-key-scans-default.test | 2 +-
.../queries/PlannerTest/partition-key-scans.test | 2 +-
.../queries/PlannerTest/tpcds-processing-cost.test | 16 +--
.../queries/PlannerTest/tpcds/tpcds-q14a.test | 16 +--
.../queries/PlannerTest/tpcds/tpcds-q18.test | 16 +--
.../queries/PlannerTest/tpcds/tpcds-q22.test | 16 +--
.../queries/PlannerTest/tpcds/tpcds-q27.test | 6 +-
.../queries/PlannerTest/tpcds/tpcds-q67.test | 16 +--
.../queries/PlannerTest/tpcds/tpcds-q98.test | 2 +-
11 files changed, 237 insertions(+), 74 deletions(-)
diff --git a/fe/src/main/java/org/apache/impala/planner/AggregationNode.java
b/fe/src/main/java/org/apache/impala/planner/AggregationNode.java
index 2b095f02c..641d63e9c 100644
--- a/fe/src/main/java/org/apache/impala/planner/AggregationNode.java
+++ b/fe/src/main/java/org/apache/impala/planner/AggregationNode.java
@@ -21,6 +21,8 @@ import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
+import javax.annotation.Nullable;
+
import org.apache.impala.analysis.AggregateInfo;
import org.apache.impala.analysis.Analyzer;
import org.apache.impala.analysis.CaseExpr;
@@ -91,6 +93,18 @@ public class AggregationNode extends PlanNode {
// The output is from a non-correlated scalar subquery returning at most one
value.
protected boolean isNonCorrelatedScalarSubquery_ = false;
+ // The aggregate input cardinality of this node.
+ // Initialized in computeStats(). Depending on the aggPhase_ of this node
and value of
+ // multiAggInfo_.getIsGroupingSet(), it can be initialized with either the
input
+ // cardinality of the related FIRST phase aggregation or this node's
immediate input
+ // cardinality (this child's cardinality).
+ private long aggInputCardinality_ = -1;
+
+ // Hold the estimated number of groups that will be produced by each
aggregation class.
+ // Initialized in computeStats(). May contain -1 if an aggregation class num
group
+ // can not be estimated.
+ private List<Long> aggClassNumGroups_;
+
public AggregationNode(
PlanNodeId id, PlanNode input, MultiAggregateInfo multiAggInfo, AggPhase
aggPhase) {
super(id, "AGGREGATE");
@@ -224,17 +238,25 @@ public class AggregationNode extends PlanNode {
// TODO: IMPALA-2945: this doesn't correctly take into account duplicate
keys on
// multiple nodes in a pre-aggregation.
cardinality_ = 0;
+ aggInputCardinality_ = getAggInputCardinality();
+ Preconditions.checkState(aggInputCardinality_ >= -1, aggInputCardinality_);
+ boolean unknownEstimate = false;
+ aggClassNumGroups_ = Lists.newArrayList();
for (AggregateInfo aggInfo : aggInfos_) {
// Compute the cardinality for this set of grouping exprs.
long numGroups = estimateNumGroups(aggInfo);
Preconditions.checkState(numGroups >= -1, numGroups);
+ aggClassNumGroups_.add(numGroups);
if (numGroups == -1) {
// No estimate of the number of groups is possible, can't even make a
// conservative estimate.
- cardinality_ = -1;
- break;
+ unknownEstimate = true;
+ } else {
+ cardinality_ = checkedAdd(cardinality_, numGroups);
}
- cardinality_ = checkedAdd(cardinality_, numGroups);
+ }
+ if (unknownEstimate) {
+ cardinality_ = -1;
}
// Take conjuncts into account.
@@ -257,11 +279,10 @@ public class AggregationNode extends PlanNode {
// limit the potential overestimation. We could, in future, improve this
further
// by recognizing functional dependencies.
List<Expr> groupingExprs = aggInfo.getGroupingExprs();
- long aggInputCardinality = getAggInputCardinality();
- long numGroups = estimateNumGroups(groupingExprs, aggInputCardinality);
+ long numGroups = estimateNumGroups(groupingExprs, aggInputCardinality_);
if (LOG.isTraceEnabled()) {
- LOG.trace("Node " + id_ + " numGroups= " + numGroups + "
aggInputCardinality=" +
- aggInputCardinality + " for agg class " + aggInfo.debugString());
+ LOG.trace("Node " + id_ + " numGroups= " + numGroups + "
aggInputCardinality="
+ + aggInputCardinality_ + " for agg class " + aggInfo.debugString());
}
return numGroups;
}
@@ -293,22 +314,44 @@ public class AggregationNode extends PlanNode {
}
/**
- * Compute the input cardinality to the distributed aggregation. If this is a
- * merge aggregation, we need to find the cardinality of the input to the
- * preaggregation.
- * Return -1 if unknown.
+ * Compute the input cardinality to the distributed aggregation. This
searches down
+ * for the FIRST phase aggregation node, and returns the minimum between
input
+ * cardinality of that FIRST phase aggregation vs this node's input
cardinality.
+ * However, if this is a TRANSPOSE phase of grouping set aggregation (such
as ROLLUP),
+ * it immediately returns this node's input cardinality since output
cardinality of
+ * the whole aggregation chain can be higher than input cardinality of the
FIRST phase
+ * aggregation. Return -1 if unknown.
*/
private long getAggInputCardinality() {
- PlanNode child = getChild(0);
- if (!aggPhase_.isMerge()) return child.getCardinality();
- PlanNode preAgg;
+ long inputCardinality = getChild(0).getCardinality();
+ if (multiAggInfo_.getIsGroupingSet() && aggPhase_ == AggPhase.TRANSPOSE) {
+ return inputCardinality;
+ }
+ AggregationNode firstAgg = this;
+ while (firstAgg.getAggPhase() != AggPhase.FIRST) {
+ firstAgg = getPrevAggNode(firstAgg);
+ }
+ return Math.min(inputCardinality, firstAgg.getChild(0).getCardinality());
+ }
+
+ private AggregationNode getPrevAggNode(AggregationNode aggNode) {
+ Preconditions.checkArgument(aggNode.getAggPhase() != AggPhase.FIRST);
+ PlanNode child = aggNode.getChild(0);
if (child instanceof ExchangeNode) {
- preAgg = child.getChild(0);
- } else {
- preAgg = child;
+ child = child.getChild(0);
}
- Preconditions.checkState(preAgg instanceof AggregationNode);
- return preAgg.getChild(0).getCardinality();
+ Preconditions.checkState(child instanceof AggregationNode);
+ return (AggregationNode) child;
+ }
+
+ private @Nullable AggregationNode getPrevAggInputNode() {
+ AggregationNode prevAgg = null;
+ if (aggPhase_ != AggPhase.FIRST && aggPhase_ != AggPhase.TRANSPOSE) {
+ prevAgg = getPrevAggNode(this);
+ Preconditions.checkState(aggInfos_.size() == prevAgg.aggInfos_.size());
+ Preconditions.checkState(aggInfos_.size() ==
prevAgg.aggClassNumGroups_.size());
+ }
+ return prevAgg;
}
/**
@@ -506,12 +549,30 @@ public class AggregationNode extends PlanNode {
return output;
}
+ private long getAggClassNumGroup(
+ @Nullable AggregationNode prevAgg, AggregateInfo aggInfo) {
+ if (prevAgg == null) {
+ return aggInputCardinality_;
+ } else {
+ // This aggInfo should be in-line with aggInfo of prevAgg
+ // (sizes are validated in getPrevAggInputNode).
+ int aggIdx = aggInfos_.indexOf(aggInfo);
+ long aggClassNumGroup = prevAgg.aggClassNumGroups_.get(aggIdx);
+ if (aggClassNumGroup <= -1) {
+ return aggInputCardinality_;
+ } else {
+ return Math.min(aggInputCardinality_, aggClassNumGroup);
+ }
+ }
+ }
+
@Override
public void computeProcessingCost(TQueryOptions queryOptions) {
processingCost_ = ProcessingCost.zero();
+ AggregationNode prevAgg = getPrevAggInputNode();
for (AggregateInfo aggInfo : aggInfos_) {
- ProcessingCost aggCost =
- aggInfo.computeProcessingCost(getDisplayLabel(),
getChild(0).getCardinality());
+ ProcessingCost aggCost = aggInfo.computeProcessingCost(
+ getDisplayLabel(), getAggClassNumGroup(prevAgg, aggInfo));
processingCost_ = ProcessingCost.sumCost(processingCost_, aggCost);
}
}
@@ -520,8 +581,10 @@ public class AggregationNode extends PlanNode {
public void computeNodeResourceProfile(TQueryOptions queryOptions) {
resourceProfiles_ = Lists.newArrayListWithCapacity(aggInfos_.size());
resourceProfiles_.clear();
+ AggregationNode prevAgg = getPrevAggInputNode();
for (AggregateInfo aggInfo : aggInfos_) {
- resourceProfiles_.add(computeAggClassResourceProfile(queryOptions,
aggInfo));
+ resourceProfiles_.add(computeAggClassResourceProfile(
+ queryOptions, aggInfo, getAggClassNumGroup(prevAgg, aggInfo)));
}
if (aggInfos_.size() == 1) {
nodeResourceProfile_ = resourceProfiles_.get(0);
@@ -534,7 +597,7 @@ public class AggregationNode extends PlanNode {
}
private ResourceProfile computeAggClassResourceProfile(
- TQueryOptions queryOptions, AggregateInfo aggInfo) {
+ TQueryOptions queryOptions, AggregateInfo aggInfo, long
inputCardinality) {
Preconditions.checkNotNull(
fragment_, "PlanNode must be placed into a fragment before calling
this method.");
long perInstanceCardinality =
fragment_.getPerInstanceNdv(aggInfo.getGroupingExprs());
@@ -543,8 +606,8 @@ public class AggregationNode extends PlanNode {
if (perInstanceCardinality == -1) {
perInstanceMemEstimate = DEFAULT_PER_INSTANCE_MEM;
} else {
- // Per-instance cardinality cannot be greater than the total input
cardinality.
- long inputCardinality = getChild(0).getCardinality();
+ // Per-instance cardinality cannot be greater than the total input
cardinality
+ // going into this aggregation class.
if (inputCardinality != -1) {
// Calculate the input cardinality distributed across fragment
instances.
long numInstances = fragment_.getNumInstances();
@@ -555,12 +618,12 @@ public class AggregationNode extends PlanNode {
// multiple fragment instances.
// This number was derived using empirical analysis of real-world
// and benchmark (tpch, tpcds) queries.
- perInstanceInputCardinality =
- (long) Math.ceil((inputCardinality / numInstances) *
DEFAULT_SKEW_FACTOR);
+ perInstanceInputCardinality = (long) Math.ceil(
+ ((double) inputCardinality / numInstances) *
DEFAULT_SKEW_FACTOR);
} else {
// The data is distributed through hash, it will be more balanced.
perInstanceInputCardinality =
- (long) Math.ceil(inputCardinality / numInstances);
+ (long) Math.ceil((double) inputCardinality / numInstances);
}
} else {
// When numInstances is 1 or unknown(-1),
perInstanceInputCardinality is the
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/card-agg.test
b/testdata/workloads/functional-planner/queries/PlannerTest/card-agg.test
index 7576e037a..6d4f58a78 100644
--- a/testdata/workloads/functional-planner/queries/PlannerTest/card-agg.test
+++ b/testdata/workloads/functional-planner/queries/PlannerTest/card-agg.test
@@ -311,6 +311,106 @@ PLAN-ROOT SINK
HDFS partitions=1/1 files=1 size=718.94MB
row-size=62B cardinality=6.00M
====
+# Similar like test above, but with per-class cardinality estimate higher than
+# than aggregation input cardinality.
+select l_partkey, l_suppkey, count(distinct l_quantity), count(distinct
l_shipmode), sum(1)
+from tpch_parquet.lineitem
+group by l_partkey, l_suppkey
+---- PLAN
+PLAN-ROOT SINK
+|
+03:AGGREGATE [FINALIZE]
+| output: aggif(valid_tid(2,4,5) = 2, count(l_quantity)),
aggif(valid_tid(2,4,5) = 4, count(l_shipmode)), aggif(valid_tid(2,4,5) = 5,
sum(1))
+| group by: CASE valid_tid(2,4,5) WHEN 2 THEN l_partkey WHEN 4 THEN l_partkey
WHEN 5 THEN l_partkey END, CASE valid_tid(2,4,5) WHEN 2 THEN l_suppkey WHEN 4
THEN l_suppkey WHEN 5 THEN l_suppkey END
+| row-size=40B cardinality=6.00M
+|
+02:AGGREGATE [FINALIZE]
+| Class 0
+| output: count(l_quantity)
+| group by: l_partkey, l_suppkey
+| Class 1
+| output: count(l_shipmode)
+| group by: l_partkey, l_suppkey
+| Class 2
+| output: sum:merge(1)
+| group by: l_partkey, l_suppkey
+| row-size=72B cardinality=18.00M
+|
+01:AGGREGATE
+| Class 0
+| group by: l_partkey, l_suppkey, l_quantity
+| Class 1
+| group by: l_partkey, l_suppkey, l_shipmode
+| Class 2
+| output: sum(1)
+| group by: l_partkey, l_suppkey
+| row-size=80B cardinality=18.00M
+|
+00:SCAN HDFS [tpch_parquet.lineitem]
+ HDFS partitions=1/1 files=3 size=194.00MB
+ row-size=40B cardinality=6.00M
+---- DISTRIBUTEDPLAN
+PLAN-ROOT SINK
+|
+08:EXCHANGE [UNPARTITIONED]
+|
+03:AGGREGATE [FINALIZE]
+| output: aggif(valid_tid(2,4,5) = 2, count(l_quantity)),
aggif(valid_tid(2,4,5) = 4, count(l_shipmode)), aggif(valid_tid(2,4,5) = 5,
sum(1))
+| group by: CASE valid_tid(2,4,5) WHEN 2 THEN l_partkey WHEN 4 THEN l_partkey
WHEN 5 THEN l_partkey END, CASE valid_tid(2,4,5) WHEN 2 THEN l_suppkey WHEN 4
THEN l_suppkey WHEN 5 THEN l_suppkey END
+| row-size=40B cardinality=6.00M
+|
+07:AGGREGATE [FINALIZE]
+| Class 0
+| output: count:merge(l_quantity)
+| group by: l_partkey, l_suppkey
+| Class 1
+| output: count:merge(l_shipmode)
+| group by: l_partkey, l_suppkey
+| Class 2
+| output: sum:merge(1)
+| group by: l_partkey, l_suppkey
+| row-size=72B cardinality=18.00M
+|
+06:EXCHANGE [HASH(CASE valid_tid(2,4,5) WHEN 2 THEN murmur_hash(l_partkey)
WHEN 4 THEN murmur_hash(l_partkey) WHEN 5 THEN murmur_hash(l_partkey) END,CASE
valid_tid(2,4,5) WHEN 2 THEN murmur_hash(l_suppkey) WHEN 4 THEN
murmur_hash(l_suppkey) WHEN 5 THEN murmur_hash(l_suppkey) END)]
+|
+02:AGGREGATE [STREAMING]
+| Class 0
+| output: count(l_quantity)
+| group by: l_partkey, l_suppkey
+| Class 1
+| output: count(l_shipmode)
+| group by: l_partkey, l_suppkey
+| Class 2
+| output: sum:merge(1)
+| group by: l_partkey, l_suppkey
+| row-size=72B cardinality=18.00M
+|
+05:AGGREGATE
+| Class 0
+| group by: l_partkey, l_suppkey, l_quantity
+| Class 1
+| group by: l_partkey, l_suppkey, l_shipmode
+| Class 2
+| output: sum:merge(1)
+| group by: l_partkey, l_suppkey
+| row-size=80B cardinality=18.00M
+|
+04:EXCHANGE [HASH(CASE valid_tid(1,3,5) WHEN 1 THEN murmur_hash(l_partkey)
WHEN 3 THEN murmur_hash(l_partkey) WHEN 5 THEN murmur_hash(l_partkey) END,CASE
valid_tid(1,3,5) WHEN 1 THEN murmur_hash(l_suppkey) WHEN 3 THEN
murmur_hash(l_suppkey) WHEN 5 THEN murmur_hash(l_suppkey) END,CASE
valid_tid(1,3,5) WHEN 1 THEN murmur_hash(l_quantity) WHEN 3 THEN
murmur_hash(l_shipmode) WHEN 5 THEN 0 END)]
+|
+01:AGGREGATE [STREAMING]
+| Class 0
+| group by: l_partkey, l_suppkey, l_quantity
+| Class 1
+| group by: l_partkey, l_suppkey, l_shipmode
+| Class 2
+| output: sum(1)
+| group by: l_partkey, l_suppkey
+| row-size=80B cardinality=18.00M
+|
+00:SCAN HDFS [tpch_parquet.lineitem]
+ HDFS partitions=1/1 files=3 size=194.00MB
+ row-size=40B cardinality=6.00M
+====
# Mixed grouping and non-grouping aggregations.
select count(distinct id), count(distinct int_col), count(*)
from functional.alltypes
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans-default.test
b/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans-default.test
index c210ae3a6..e3c1409d5 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans-default.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans-default.test
@@ -432,7 +432,7 @@ PLAN-ROOT SINK
03:AGGREGATE [FINALIZE]
| output: aggif(valid_tid(2,4) = 2, count('test')), aggif(valid_tid(2,4) = 4,
count(1234))
| group by: CASE valid_tid(2,4) WHEN 2 THEN 1234 WHEN 4 THEN 1234 END
-| row-size=18B cardinality=2
+| row-size=18B cardinality=1
|
02:AGGREGATE [FINALIZE]
| Class 0
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans.test
b/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans.test
index 232d07440..1130922eb 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/partition-key-scans.test
@@ -374,7 +374,7 @@ PLAN-ROOT SINK
03:AGGREGATE [FINALIZE]
| output: aggif(valid_tid(2,4) = 2, count('test')), aggif(valid_tid(2,4) = 4,
count(1234))
| group by: CASE valid_tid(2,4) WHEN 2 THEN 1234 WHEN 4 THEN 1234 END
-| row-size=18B cardinality=2
+| row-size=18B cardinality=1
|
02:AGGREGATE [FINALIZE]
| Class 0
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds-processing-cost.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds-processing-cost.test
index 94307dbfe..390e58354 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds-processing-cost.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds-processing-cost.test
@@ -270,8 +270,8 @@ with cross_items as
order by channel,i_brand_id,i_class_id,i_category_id
LIMIT 100
---- PARALLELPLANS
-Max Per-Host Resource Reservation: Memory=790.00MB Threads=163
-Per-Host Resource Estimates: Memory=2.42GB
+Max Per-Host Resource Reservation: Memory=716.38MB Threads=163
+Per-Host Resource Estimates: Memory=2.36GB
F80:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Instance Resources: mem-estimate=4.03MB mem-reservation=4.00MB
thread-reservation=1
| max-parallelism=1 segment-costs=[606] cpu-comparison-result=157 [max(1
(self) vs 157 (sum children))]
@@ -287,8 +287,8 @@ PLAN-ROOT SINK
| in pipelines: 129(GETNEXT)
|
F79:PLAN FRAGMENT [HASH(CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(channel) WHEN 105 THEN murmur_hash(channel) WHEN 106 THEN
murmur_hash(channel) WHEN 107 THEN murmur_hash(channel) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(i_brand_id) WHEN 105 THEN murmur_hash(i_brand_id) WHEN 106 THEN
murmur_hash(i_brand_id) WHEN 107 THEN murmur_hash(NULL) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WH [...]
-Per-Instance Resources: mem-estimate=111.05MB mem-reservation=85.12MB
thread-reservation=1
-max-parallelism=6 segment-costs=[17011772, 3936100, 400, 6]
cpu-comparison-result=157 [max(6 (self) vs 157 (sum children))]
+Per-Instance Resources: mem-estimate=77.05MB mem-reservation=48.31MB
thread-reservation=1
+max-parallelism=6 segment-costs=[3516572, 3936100, 400, 6]
cpu-comparison-result=157 [max(6 (self) vs 157 (sum children))]
129:TOP-N [LIMIT=100]
| order by: CASE valid_tid(104,105,106,107,108) WHEN 104 THEN channel WHEN
105 THEN channel WHEN 106 THEN channel WHEN 107 THEN channel WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_brand_id WHEN 105
THEN i_brand_id WHEN 106 THEN i_brand_id WHEN 107 THEN NULL WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_class_id WHEN 105
THEN i_class_id WHEN 106 THEN NULL WHEN 107 THEN NULL WHEN 108 THEN NULL END
ASC, CASE valid_tid(10 [...]
| mem-estimate=4.69KB mem-reservation=0B thread-reservation=0
@@ -318,8 +318,8 @@ max-parallelism=6 segment-costs=[17011772, 3936100, 400, 6]
cpu-comparison-resul
| Class 4
| output: sum:merge(sales), sum:merge(number_sales)
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=98.00MB mem-reservation=76.62MB thread-reservation=0
-| tuple-ids=104N,105N,106N,107N,108N row-size=240B cardinality=562.30K
cost=16869000
+| mem-estimate=64.00MB mem-reservation=39.81MB thread-reservation=0
+| tuple-ids=104N,105N,106N,107N,108N row-size=240B cardinality=562.30K
cost=3373800
| in pipelines: 215(GETNEXT), 151(OPEN), 179(OPEN), 207(OPEN)
|
214:EXCHANGE [HASH(CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(channel) WHEN 105 THEN murmur_hash(channel) WHEN 106 THEN
murmur_hash(channel) WHEN 107 THEN murmur_hash(channel) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(i_brand_id) WHEN 105 THEN murmur_hash(i_brand_id) WHEN 106 THEN
murmur_hash(i_brand_id) WHEN 107 THEN murmur_hash(NULL) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WHEN 10 [...]
@@ -8007,7 +8007,7 @@ PLAN-ROOT SINK
|
F22:PLAN FRAGMENT [HASH(CASE valid_tid(26,27,28) WHEN 26 THEN
murmur_hash(channel) WHEN 27 THEN murmur_hash(channel) WHEN 28 THEN
murmur_hash(NULL) END,CASE valid_tid(26,27,28) WHEN 26 THEN murmur_hash(id)
WHEN 27 THEN murmur_hash(NULL) WHEN 28 THEN murmur_hash(NULL) END)] hosts=3
instances=6 (adjusted from 48)
Per-Instance Resources: mem-estimate=40.00MB mem-reservation=7.75MB
thread-reservation=1
-max-parallelism=6 segment-costs=[176035, 69384, 200, 8]
cpu-comparison-result=55 [max(6 (self) vs 55 (sum children))]
+max-parallelism=6 segment-costs=[60395, 69384, 200, 8]
cpu-comparison-result=55 [max(6 (self) vs 55 (sum children))]
39:TOP-N [LIMIT=100]
| order by: CASE valid_tid(26,27,28) WHEN 26 THEN channel WHEN 27 THEN
channel WHEN 28 THEN NULL END ASC, CASE valid_tid(26,27,28) WHEN 26 THEN id
WHEN 27 THEN NULL WHEN 28 THEN NULL END ASC
| mem-estimate=7.03KB mem-reservation=0B thread-reservation=0
@@ -8032,7 +8032,7 @@ max-parallelism=6 segment-costs=[176035, 69384, 200, 8]
cpu-comparison-result=55
| output: sum:merge(sales), sum:merge(`returns`), sum:merge(profit)
| group by: NULL, NULL
| mem-estimate=30.00MB mem-reservation=5.81MB thread-reservation=0
-| tuple-ids=26N,27N,28N row-size=216B cardinality=11.56K cost=173460
+| tuple-ids=26N,27N,28N row-size=216B cardinality=11.56K cost=57820
| in pipelines: 62(GETNEXT), 46(OPEN), 53(OPEN), 60(OPEN)
|
61:EXCHANGE [HASH(CASE valid_tid(26,27,28) WHEN 26 THEN murmur_hash(channel)
WHEN 27 THEN murmur_hash(channel) WHEN 28 THEN murmur_hash(NULL) END,CASE
valid_tid(26,27,28) WHEN 26 THEN murmur_hash(id) WHEN 27 THEN murmur_hash(NULL)
WHEN 28 THEN murmur_hash(NULL) END)]
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14a.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14a.test
index 6d4222ac8..a3d26ad83 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14a.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q14a.test
@@ -1295,8 +1295,8 @@ PLAN-ROOT SINK
tuple-ids=0 row-size=20B cardinality=2.88M
in pipelines: 01(GETNEXT)
---- DISTRIBUTEDPLAN
-Max Per-Host Resource Reservation: Memory=427.00MB Threads=120
-Per-Host Resource Estimates: Memory=3.56GB
+Max Per-Host Resource Reservation: Memory=424.19MB Threads=120
+Per-Host Resource Estimates: Memory=3.54GB
F80:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Host Resources: mem-estimate=4.02MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -1311,7 +1311,7 @@ PLAN-ROOT SINK
| in pipelines: 129(GETNEXT)
|
F79:PLAN FRAGMENT [HASH(CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(channel) WHEN 105 THEN murmur_hash(channel) WHEN 106 THEN
murmur_hash(channel) WHEN 107 THEN murmur_hash(channel) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(i_brand_id) WHEN 105 THEN murmur_hash(i_brand_id) WHEN 106 THEN
murmur_hash(i_brand_id) WHEN 107 THEN murmur_hash(NULL) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WH [...]
-Per-Host Resources: mem-estimate=137.09MB mem-reservation=93.62MB
thread-reservation=1
+Per-Host Resources: mem-estimate=115.00MB mem-reservation=90.81MB
thread-reservation=1
129:TOP-N [LIMIT=100]
| order by: CASE valid_tid(104,105,106,107,108) WHEN 104 THEN channel WHEN
105 THEN channel WHEN 106 THEN channel WHEN 107 THEN channel WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_brand_id WHEN 105
THEN i_brand_id WHEN 106 THEN i_brand_id WHEN 107 THEN NULL WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_class_id WHEN 105
THEN i_class_id WHEN 106 THEN NULL WHEN 107 THEN NULL WHEN 108 THEN NULL END
ASC, CASE valid_tid(10 [...]
| mem-estimate=4.69KB mem-reservation=0B thread-reservation=0
@@ -1341,7 +1341,7 @@ Per-Host Resources: mem-estimate=137.09MB
mem-reservation=93.62MB thread-reserva
| Class 4
| output: sum:merge(sales), sum:merge(number_sales)
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=120.09MB mem-reservation=76.62MB thread-reservation=0
+| mem-estimate=98.00MB mem-reservation=73.81MB thread-reservation=0
| tuple-ids=104N,105N,106N,107N,108N row-size=240B cardinality=562.30K
| in pipelines: 215(GETNEXT), 151(OPEN), 179(OPEN), 207(OPEN)
|
@@ -3098,8 +3098,8 @@ Per-Host Resources: mem-estimate=67.37MB
mem-reservation=12.81MB thread-reservat
tuple-ids=0 row-size=20B cardinality=2.88M
in pipelines: 01(GETNEXT)
---- PARALLELPLANS
-Max Per-Host Resource Reservation: Memory=662.12MB Threads=141
-Per-Host Resource Estimates: Memory=2.28GB
+Max Per-Host Resource Reservation: Memory=588.50MB Threads=141
+Per-Host Resource Estimates: Memory=2.22GB
F80:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Instance Resources: mem-estimate=4.03MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -3114,7 +3114,7 @@ PLAN-ROOT SINK
| in pipelines: 129(GETNEXT)
|
F79:PLAN FRAGMENT [HASH(CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(channel) WHEN 105 THEN murmur_hash(channel) WHEN 106 THEN
murmur_hash(channel) WHEN 107 THEN murmur_hash(channel) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WHEN 104 THEN
murmur_hash(i_brand_id) WHEN 105 THEN murmur_hash(i_brand_id) WHEN 106 THEN
murmur_hash(i_brand_id) WHEN 107 THEN murmur_hash(NULL) WHEN 108 THEN
murmur_hash(NULL) END,CASE valid_tid(104,105,106,107,108) WH [...]
-Per-Instance Resources: mem-estimate=109.52MB mem-reservation=85.12MB
thread-reservation=1
+Per-Instance Resources: mem-estimate=75.52MB mem-reservation=48.31MB
thread-reservation=1
129:TOP-N [LIMIT=100]
| order by: CASE valid_tid(104,105,106,107,108) WHEN 104 THEN channel WHEN
105 THEN channel WHEN 106 THEN channel WHEN 107 THEN channel WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_brand_id WHEN 105
THEN i_brand_id WHEN 106 THEN i_brand_id WHEN 107 THEN NULL WHEN 108 THEN NULL
END ASC, CASE valid_tid(104,105,106,107,108) WHEN 104 THEN i_class_id WHEN 105
THEN i_class_id WHEN 106 THEN NULL WHEN 107 THEN NULL WHEN 108 THEN NULL END
ASC, CASE valid_tid(10 [...]
| mem-estimate=4.69KB mem-reservation=0B thread-reservation=0
@@ -3144,7 +3144,7 @@ Per-Instance Resources: mem-estimate=109.52MB
mem-reservation=85.12MB thread-res
| Class 4
| output: sum:merge(sales), sum:merge(number_sales)
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=98.00MB mem-reservation=76.62MB thread-reservation=0
+| mem-estimate=64.00MB mem-reservation=39.81MB thread-reservation=0
| tuple-ids=104N,105N,106N,107N,108N row-size=240B cardinality=562.30K
| in pipelines: 215(GETNEXT), 151(OPEN), 179(OPEN), 207(OPEN)
|
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q18.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q18.test
index a2502d510..a189fb254 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q18.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q18.test
@@ -206,8 +206,8 @@ PLAN-ROOT SINK
tuple-ids=2 row-size=4B cardinality=1.92M
in pipelines: 02(GETNEXT)
---- DISTRIBUTEDPLAN
-Max Per-Host Resource Reservation: Memory=194.25MB Threads=18
-Per-Host Resource Estimates: Memory=888MB
+Max Per-Host Resource Reservation: Memory=166.00MB Threads=18
+Per-Host Resource Estimates: Memory=874MB
F10:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Host Resources: mem-estimate=4.03MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -222,7 +222,7 @@ PLAN-ROOT SINK
| in pipelines: 15(GETNEXT)
|
F09:PLAN FRAGMENT [HASH(CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(i_item_id) WHEN 9 THEN murmur_hash(i_item_id) WHEN 11 THEN
murmur_hash(i_item_id) WHEN 13 THEN murmur_hash(i_item_id) WHEN 15 THEN
murmur_hash(NULL) END,CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(ca_country) WHEN 9 THEN murmur_hash(ca_country) WHEN 11 THEN
murmur_hash(ca_country) WHEN 13 THEN murmur_hash(NULL) WHEN 15 THEN
murmur_hash(NULL) END,CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(ca_st [...]
-Per-Host Resources: mem-estimate=75.94MB mem-reservation=57.69MB
thread-reservation=1
+Per-Host Resources: mem-estimate=61.94MB mem-reservation=29.44MB
thread-reservation=1
15:TOP-N [LIMIT=100]
| order by: CASE valid_tid(16,8,10,12,14) WHEN 8 THEN ca_country WHEN 10 THEN
ca_country WHEN 12 THEN ca_country WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC,
CASE valid_tid(16,8,10,12,14) WHEN 8 THEN ca_state WHEN 10 THEN ca_state WHEN
12 THEN NULL WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC, CASE
valid_tid(16,8,10,12,14) WHEN 8 THEN ca_county WHEN 10 THEN NULL WHEN 12 THEN
NULL WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC, CASE valid_tid(16,8,10,12,14)
WHEN 8 THEN i_item_id WHEN 10 THEN [...]
| mem-estimate=10.16KB mem-reservation=0B thread-reservation=0
@@ -252,7 +252,7 @@ Per-Host Resources: mem-estimate=75.94MB
mem-reservation=57.69MB thread-reservat
| Class 4
| output: avg:merge(CAST(cs_quantity AS DECIMAL(12,2))),
avg:merge(CAST(cs_list_price AS DECIMAL(12,2))), avg:merge(CAST(cs_coupon_amt
AS DECIMAL(12,2))), avg:merge(CAST(cs_sales_price AS DECIMAL(12,2))),
avg:merge(CAST(cs_net_profit AS DECIMAL(12,2))), avg:merge(CAST(c_birth_year AS
DECIMAL(12,2))), avg:merge(CAST(cd1.cd_dep_count AS DECIMAL(12,2)))
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=64.00MB mem-reservation=52.94MB thread-reservation=0
+| mem-estimate=50.00MB mem-reservation=24.69MB thread-reservation=0
| tuple-ids=8N,10N,12N,14N,16N row-size=641B cardinality=75.61K
| in pipelines: 25(GETNEXT), 00(OPEN)
|
@@ -471,8 +471,8 @@ Per-Host Resources: mem-estimate=399.42MB
mem-reservation=29.62MB thread-reserva
tuple-ids=0 row-size=40B cardinality=1.44M
in pipelines: 00(GETNEXT)
---- PARALLELPLANS
-Max Per-Host Resource Reservation: Memory=206.81MB Threads=17
-Per-Host Resource Estimates: Memory=384MB
+Max Per-Host Resource Reservation: Memory=178.56MB Threads=17
+Per-Host Resource Estimates: Memory=370MB
F10:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Instance Resources: mem-estimate=4.03MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -487,7 +487,7 @@ PLAN-ROOT SINK
| in pipelines: 15(GETNEXT)
|
F09:PLAN FRAGMENT [HASH(CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(i_item_id) WHEN 9 THEN murmur_hash(i_item_id) WHEN 11 THEN
murmur_hash(i_item_id) WHEN 13 THEN murmur_hash(i_item_id) WHEN 15 THEN
murmur_hash(NULL) END,CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(ca_country) WHEN 9 THEN murmur_hash(ca_country) WHEN 11 THEN
murmur_hash(ca_country) WHEN 13 THEN murmur_hash(NULL) WHEN 15 THEN
murmur_hash(NULL) END,CASE valid_tid(7,9,11,13,15) WHEN 7 THEN
murmur_hash(ca_st [...]
-Per-Instance Resources: mem-estimate=75.94MB mem-reservation=57.69MB
thread-reservation=1
+Per-Instance Resources: mem-estimate=61.94MB mem-reservation=29.44MB
thread-reservation=1
15:TOP-N [LIMIT=100]
| order by: CASE valid_tid(16,8,10,12,14) WHEN 8 THEN ca_country WHEN 10 THEN
ca_country WHEN 12 THEN ca_country WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC,
CASE valid_tid(16,8,10,12,14) WHEN 8 THEN ca_state WHEN 10 THEN ca_state WHEN
12 THEN NULL WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC, CASE
valid_tid(16,8,10,12,14) WHEN 8 THEN ca_county WHEN 10 THEN NULL WHEN 12 THEN
NULL WHEN 14 THEN NULL WHEN 16 THEN NULL END ASC, CASE valid_tid(16,8,10,12,14)
WHEN 8 THEN i_item_id WHEN 10 THEN [...]
| mem-estimate=10.16KB mem-reservation=0B thread-reservation=0
@@ -517,7 +517,7 @@ Per-Instance Resources: mem-estimate=75.94MB
mem-reservation=57.69MB thread-rese
| Class 4
| output: avg:merge(CAST(cs_quantity AS DECIMAL(12,2))),
avg:merge(CAST(cs_list_price AS DECIMAL(12,2))), avg:merge(CAST(cs_coupon_amt
AS DECIMAL(12,2))), avg:merge(CAST(cs_sales_price AS DECIMAL(12,2))),
avg:merge(CAST(cs_net_profit AS DECIMAL(12,2))), avg:merge(CAST(c_birth_year AS
DECIMAL(12,2))), avg:merge(CAST(cd1.cd_dep_count AS DECIMAL(12,2)))
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=64.00MB mem-reservation=52.94MB thread-reservation=0
+| mem-estimate=50.00MB mem-reservation=24.69MB thread-reservation=0
| tuple-ids=8N,10N,12N,14N,16N row-size=641B cardinality=75.61K
| in pipelines: 25(GETNEXT), 00(OPEN)
|
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q22.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q22.test
index cd555434a..1288f9826 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q22.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q22.test
@@ -128,8 +128,8 @@ PLAN-ROOT SINK
tuple-ids=0 row-size=20B cardinality=11.74M
in pipelines: 00(GETNEXT)
---- DISTRIBUTEDPLAN
-Max Per-Host Resource Reservation: Memory=292.57MB Threads=10
-Per-Host Resource Estimates: Memory=26.12GB
+Max Per-Host Resource Reservation: Memory=288.82MB Threads=10
+Per-Host Resource Estimates: Memory=13.96GB
F05:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Host Resources: mem-estimate=4.02MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -144,7 +144,7 @@ PLAN-ROOT SINK
| in pipelines: 09(GETNEXT)
|
F04:PLAN FRAGMENT [HASH(CASE valid_tid(4,6,8,10,12) WHEN 4 THEN
murmur_hash(i_product_name) WHEN 6 THEN murmur_hash(i_product_name) WHEN 8 THEN
murmur_hash(i_product_name) WHEN 10 THEN murmur_hash(i_product_name) WHEN 12
THEN murmur_hash(NULL) END,CASE valid_tid(4,6,8,10,12) WHEN 4 THEN
murmur_hash(i_brand) WHEN 6 THEN murmur_hash(i_brand) WHEN 8 THEN
murmur_hash(i_brand) WHEN 10 THEN murmur_hash(NULL) WHEN 12 THEN
murmur_hash(NULL) END,CASE valid_tid(4,6,8,10,12) WHEN 4 THEN murmur_hash [...]
-Per-Host Resources: mem-estimate=20.49GB mem-reservation=146.44MB
thread-reservation=1
+Per-Host Resources: mem-estimate=8.33GB mem-reservation=142.69MB
thread-reservation=1
09:TOP-N [LIMIT=100]
| order by: aggif(valid_tid(5,7,9,11,13) IN (5, 7, 9, 11, 13), CASE
valid_tid(5,7,9,11,13) WHEN 5 THEN avg(inv_quantity_on_hand) WHEN 7 THEN
avg(inv_quantity_on_hand) WHEN 9 THEN avg(inv_quantity_on_hand) WHEN 11 THEN
avg(inv_quantity_on_hand) WHEN 13 THEN avg(inv_quantity_on_hand) END) ASC, CASE
valid_tid(5,7,9,11,13) WHEN 5 THEN i_product_name WHEN 7 THEN i_product_name
WHEN 9 THEN i_product_name WHEN 11 THEN i_product_name WHEN 13 THEN NULL END
ASC, CASE valid_tid(5,7,9,11,13) WHEN 5 [...]
| mem-estimate=5.47KB mem-reservation=0B thread-reservation=0
@@ -174,7 +174,7 @@ Per-Host Resources: mem-estimate=20.49GB
mem-reservation=146.44MB thread-reserva
| Class 4
| output: avg:merge(inv_quantity_on_hand)
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=19.30GB mem-reservation=112.44MB thread-reservation=0
+| mem-estimate=7.14GB mem-reservation=108.69MB thread-reservation=0
| tuple-ids=5N,7N,9N,11N,13N row-size=422B cardinality=35.25M
| in pipelines: 14(GETNEXT), 00(OPEN)
|
@@ -294,8 +294,8 @@ Per-Host Resources: mem-estimate=5.50GB
mem-reservation=140.62MB thread-reservat
tuple-ids=0 row-size=20B cardinality=11.74M
in pipelines: 00(GETNEXT)
---- PARALLELPLANS
-Max Per-Host Resource Reservation: Memory=304.20MB Threads=9
-Per-Host Resource Estimates: Memory=25.96GB
+Max Per-Host Resource Reservation: Memory=300.45MB Threads=9
+Per-Host Resource Estimates: Memory=13.80GB
F05:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Instance Resources: mem-estimate=4.02MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -310,7 +310,7 @@ PLAN-ROOT SINK
| in pipelines: 09(GETNEXT)
|
F04:PLAN FRAGMENT [HASH(CASE valid_tid(4,6,8,10,12) WHEN 4 THEN
murmur_hash(i_product_name) WHEN 6 THEN murmur_hash(i_product_name) WHEN 8 THEN
murmur_hash(i_product_name) WHEN 10 THEN murmur_hash(i_product_name) WHEN 12
THEN murmur_hash(NULL) END,CASE valid_tid(4,6,8,10,12) WHEN 4 THEN
murmur_hash(i_brand) WHEN 6 THEN murmur_hash(i_brand) WHEN 8 THEN
murmur_hash(i_brand) WHEN 10 THEN murmur_hash(NULL) WHEN 12 THEN
murmur_hash(NULL) END,CASE valid_tid(4,6,8,10,12) WHEN 4 THEN murmur_hash [...]
-Per-Instance Resources: mem-estimate=20.49GB mem-reservation=146.44MB
thread-reservation=1
+Per-Instance Resources: mem-estimate=8.33GB mem-reservation=142.69MB
thread-reservation=1
09:TOP-N [LIMIT=100]
| order by: aggif(valid_tid(5,7,9,11,13) IN (5, 7, 9, 11, 13), CASE
valid_tid(5,7,9,11,13) WHEN 5 THEN avg(inv_quantity_on_hand) WHEN 7 THEN
avg(inv_quantity_on_hand) WHEN 9 THEN avg(inv_quantity_on_hand) WHEN 11 THEN
avg(inv_quantity_on_hand) WHEN 13 THEN avg(inv_quantity_on_hand) END) ASC, CASE
valid_tid(5,7,9,11,13) WHEN 5 THEN i_product_name WHEN 7 THEN i_product_name
WHEN 9 THEN i_product_name WHEN 11 THEN i_product_name WHEN 13 THEN NULL END
ASC, CASE valid_tid(5,7,9,11,13) WHEN 5 [...]
| mem-estimate=5.47KB mem-reservation=0B thread-reservation=0
@@ -340,7 +340,7 @@ Per-Instance Resources: mem-estimate=20.49GB
mem-reservation=146.44MB thread-res
| Class 4
| output: avg:merge(inv_quantity_on_hand)
| group by: NULL, NULL, NULL, NULL
-| mem-estimate=19.30GB mem-reservation=112.44MB thread-reservation=0
+| mem-estimate=7.14GB mem-reservation=108.69MB thread-reservation=0
| tuple-ids=5N,7N,9N,11N,13N row-size=422B cardinality=35.25M
| in pipelines: 14(GETNEXT), 00(OPEN)
|
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q27.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q27.test
index bcde921c2..5a75e5dc9 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q27.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q27.test
@@ -148,7 +148,7 @@ PLAN-ROOT SINK
tuple-ids=0 row-size=36B cardinality=2.88M
in pipelines: 00(GETNEXT)
---- DISTRIBUTEDPLAN
-Max Per-Host Resource Reservation: Memory=52.95MB Threads=12
+Max Per-Host Resource Reservation: Memory=51.08MB Threads=12
Per-Host Resource Estimates: Memory=370MB
F06:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Host Resources: mem-estimate=4.02MB mem-reservation=4.00MB
thread-reservation=1
@@ -164,7 +164,7 @@ PLAN-ROOT SINK
| in pipelines: 11(GETNEXT)
|
F05:PLAN FRAGMENT [HASH(CASE valid_tid(5,7,9) WHEN 5 THEN
murmur_hash(i_item_id) WHEN 7 THEN murmur_hash(i_item_id) WHEN 9 THEN
murmur_hash(NULL) END,CASE valid_tid(5,7,9) WHEN 5 THEN murmur_hash(s_state)
WHEN 7 THEN murmur_hash(NULL) WHEN 9 THEN murmur_hash(NULL) END)] hosts=3
instances=3
-Per-Host Resources: mem-estimate=40.00MB mem-reservation=9.62MB
thread-reservation=1
+Per-Host Resources: mem-estimate=40.00MB mem-reservation=7.75MB
thread-reservation=1
11:TOP-N [LIMIT=100]
| order by: CASE valid_tid(6,8,10) WHEN 6 THEN i_item_id WHEN 8 THEN
i_item_id WHEN 10 THEN NULL END ASC, CASE valid_tid(6,8,10) WHEN 6 THEN s_state
WHEN 8 THEN NULL WHEN 10 THEN NULL END ASC
| mem-estimate=5.57KB mem-reservation=0B thread-reservation=0
@@ -188,7 +188,7 @@ Per-Host Resources: mem-estimate=40.00MB
mem-reservation=9.62MB thread-reservati
| Class 2
| output: avg:merge(ss_quantity), avg:merge(ss_list_price),
avg:merge(ss_coupon_amt), avg:merge(ss_sales_price)
| group by: NULL, NULL
-| mem-estimate=30.00MB mem-reservation=7.69MB thread-reservation=0
+| mem-estimate=30.00MB mem-reservation=5.81MB thread-reservation=0
| tuple-ids=6N,8N,10N row-size=202B cardinality=17.71K
| in pipelines: 17(GETNEXT), 00(OPEN)
|
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q67.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q67.test
index 4697fdd99..5de0877a8 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q67.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q67.test
@@ -186,8 +186,8 @@ PLAN-ROOT SINK
tuple-ids=0 row-size=24B cardinality=2.88M
in pipelines: 00(GETNEXT)
---- DISTRIBUTEDPLAN
-Max Per-Host Resource Reservation: Memory=478.39MB Threads=11
-Per-Host Resource Estimates: Memory=31.93GB
+Max Per-Host Resource Reservation: Memory=477.45MB Threads=11
+Per-Host Resource Estimates: Memory=10.45GB
F06:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Host Resources: mem-estimate=4.03MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -237,7 +237,7 @@ Per-Host Resources: mem-estimate=4.08MB
mem-reservation=4.00MB thread-reservatio
| in pipelines: 09(GETNEXT)
|
F04:PLAN FRAGMENT [HASH(CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN
murmur_hash(i_category) WHEN 5 THEN murmur_hash(i_category) WHEN 6 THEN
murmur_hash(i_category) WHEN 7 THEN murmur_hash(i_category) WHEN 8 THEN
murmur_hash(i_category) WHEN 9 THEN murmur_hash(i_category) WHEN 10 THEN
murmur_hash(i_category) WHEN 11 THEN murmur_hash(i_category) WHEN 12 THEN
murmur_hash(NULL) END,CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN
murmur_hash(i_class) WHEN 5 THEN murmur_hash(i_class) WHE [...]
-Per-Host Resources: mem-estimate=27.19GB mem-reservation=244.75MB
thread-reservation=1
+Per-Host Resources: mem-estimate=5.71GB mem-reservation=243.81MB
thread-reservation=1
09:TOP-N
| order by: CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN i_category WHEN
5 THEN i_category WHEN 6 THEN i_category WHEN 7 THEN i_category WHEN 8 THEN
i_category WHEN 9 THEN i_category WHEN 10 THEN i_category WHEN 11 THEN
i_category WHEN 12 THEN NULL END ASC NULLS LAST,
aggif(valid_tid(4,5,6,7,8,9,10,11,12) IN (4, 5, 6, 7, 8, 9, 10, 11, 12), CASE
valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN sum(coalesce(ss_sales_price *
ss_quantity, 0)) WHEN 5 THEN sum(coalesce(ss_sales_price * ss_quan [...]
| limit with ties: 200
@@ -281,7 +281,7 @@ Per-Host Resources: mem-estimate=27.19GB
mem-reservation=244.75MB thread-reserva
| Class 8
| output: sum:merge(coalesce(ss_sales_price * ss_quantity, 0))
| group by: NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL
-| mem-estimate=26.71GB mem-reservation=210.75MB thread-reservation=0
+| mem-estimate=5.22GB mem-reservation=209.81MB thread-reservation=0
| tuple-ids=4N,5N,6N,7N,8N,9N,10N,11N,12N row-size=1.07KB cardinality=15.09M
| in pipelines: 17(GETNEXT), 00(OPEN)
|
@@ -414,8 +414,8 @@ Per-Host Resources: mem-estimate=4.54GB
mem-reservation=223.62MB thread-reservat
tuple-ids=0 row-size=24B cardinality=2.88M
in pipelines: 00(GETNEXT)
---- PARALLELPLANS
-Max Per-Host Resource Reservation: Memory=950.77MB Threads=13
-Per-Host Resource Estimates: Memory=32.62GB
+Max Per-Host Resource Reservation: Memory=948.89MB Threads=13
+Per-Host Resource Estimates: Memory=10.43GB
F06:PLAN FRAGMENT [UNPARTITIONED] hosts=1 instances=1
| Per-Instance Resources: mem-estimate=4.06MB mem-reservation=4.00MB
thread-reservation=1
PLAN-ROOT SINK
@@ -465,7 +465,7 @@ Per-Instance Resources: mem-estimate=4.13MB
mem-reservation=4.00MB thread-reserv
| in pipelines: 09(GETNEXT)
|
F04:PLAN FRAGMENT [HASH(CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN
murmur_hash(i_category) WHEN 5 THEN murmur_hash(i_category) WHEN 6 THEN
murmur_hash(i_category) WHEN 7 THEN murmur_hash(i_category) WHEN 8 THEN
murmur_hash(i_category) WHEN 9 THEN murmur_hash(i_category) WHEN 10 THEN
murmur_hash(i_category) WHEN 11 THEN murmur_hash(i_category) WHEN 12 THEN
murmur_hash(NULL) END,CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN
murmur_hash(i_class) WHEN 5 THEN murmur_hash(i_class) WHE [...]
-Per-Instance Resources: mem-estimate=13.96GB mem-reservation=244.75MB
thread-reservation=1
+Per-Instance Resources: mem-estimate=2.87GB mem-reservation=243.81MB
thread-reservation=1
09:TOP-N
| order by: CASE valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN i_category WHEN
5 THEN i_category WHEN 6 THEN i_category WHEN 7 THEN i_category WHEN 8 THEN
i_category WHEN 9 THEN i_category WHEN 10 THEN i_category WHEN 11 THEN
i_category WHEN 12 THEN NULL END ASC NULLS LAST,
aggif(valid_tid(4,5,6,7,8,9,10,11,12) IN (4, 5, 6, 7, 8, 9, 10, 11, 12), CASE
valid_tid(4,5,6,7,8,9,10,11,12) WHEN 4 THEN sum(coalesce(ss_sales_price *
ss_quantity, 0)) WHEN 5 THEN sum(coalesce(ss_sales_price * ss_quan [...]
| limit with ties: 200
@@ -509,7 +509,7 @@ Per-Instance Resources: mem-estimate=13.96GB
mem-reservation=244.75MB thread-res
| Class 8
| output: sum:merge(coalesce(ss_sales_price * ss_quantity, 0))
| group by: NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL
-| mem-estimate=13.72GB mem-reservation=210.75MB thread-reservation=0
+| mem-estimate=2.63GB mem-reservation=209.81MB thread-reservation=0
| tuple-ids=4N,5N,6N,7N,8N,9N,10N,11N,12N row-size=1.07KB cardinality=15.09M
| in pipelines: 17(GETNEXT), 00(OPEN)
|
diff --git
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q98.test
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q98.test
index 0d232e71e..c1d471054 100644
---
a/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q98.test
+++
b/testdata/workloads/functional-planner/queries/PlannerTest/tpcds/tpcds-q98.test
@@ -133,7 +133,7 @@ PLAN-ROOT SINK
| in pipelines: 08(GETNEXT)
|
F03:PLAN FRAGMENT [HASH(i_class)] hosts=3 instances=3
-Per-Host Resources: mem-estimate=116.13MB mem-reservation=50.00MB
thread-reservation=1
+Per-Host Resources: mem-estimate=116.14MB mem-reservation=50.00MB
thread-reservation=1
08:TOP-N [LIMIT=1000]
| order by: i_category ASC, i_class ASC, i_item_id ASC, i_item_desc ASC,
sum(ss_ext_sales_price) * 100 / sum(sum(ss_ext_sales_price)) ASC
| mem-estimate=209.09KB mem-reservation=0B thread-reservation=0
