Backport of:

timestamp: Thu 2011-12-01 15:12:10 +0100
Fix for Bug#13430436 PERFORMANCE DEGRADATION IN SYSBENCH ON INNODB DUE TO ICP

When running sysbench on InnoDB there is a performance degradation due
to index condition pushdown (ICP). Several of the queries in sysbench
have a WHERE condition that the optimizer uses for executing these
queries as range scans. The upper and lower limit of the range scan
will ensure that the WHERE condition is fulfilled. Still, the WHERE
condition is part of the queries' condition and if ICP is enabled the
condition will be pushed down to InnoDB as an index condition. 

Due to the range scan's upper and lower limits ensure that the WHERE
condition is fulfilled, the pushed index condition will not filter out
any records. As a result the use of ICP for these queries results in a
performance overhead for sysbench. This overhead comes from using
resources for determining the part of the condition that can be pushed
down to InnoDB and overhead in InnoDB for executing the pushed index
condition.

With the default configuration for sysbench the range scans will use
the primary key. This is a clustered index in InnoDB. Using ICP on a
clustered index provides the lowest performance benefit since the
entire record is part of the clustered index and in InnoDB it has the
highest relative overhead for executing the pushed index condition.

The fix for removing the overhead ICP introduces when running sysbench
is to disable use of ICP when the index used by the query is a
clustered index.

When WL#6061 is implemented this change should be re-evaluated.

1 files changed, 3 insertions, 3 deletions

diff --git a/mysql-test/r/innodb_mrr_cpk.result b/mysql-test/r/innodb_mrr_cpk.result
index 81536f2a43b..90f59b96e61 100644
--- a/mysql-test/r/innodb_mrr_cpk.result
+++ b/mysql-test/r/innodb_mrr_cpk.result
@@ -82,7 +82,7 @@ insert into t2 values ('a-1010=A', 1010), ('a-1030=A', 1030), ('a-1020=A', 1020)
 explain select * from t1, t2 where t1.a=t2.a and t1.b=t2.b;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	SIMPLE	t2	ALL	NULL	NULL	NULL	NULL	3	Using where
-1	SIMPLE	t1	eq_ref	PRIMARY	PRIMARY	30	test.t2.a,test.t2.b	1	Using index condition(BKA); Using join buffer (flat, BKA join); Key-ordered scan
+1	SIMPLE	t1	eq_ref	PRIMARY	PRIMARY	30	test.t2.a,test.t2.b	1	Using where; Using join buffer (flat, BKA join); Key-ordered scan
 select * from t1, t2 where t1.a=t2.a and t1.b=t2.b;
 a	b	filler	a	b
 a-1010=A	1010	filler	a-1010=A	1010
@@ -91,7 +91,7 @@ a-1030=A	1030	filler	a-1030=A	1030
 explain select * from t1, t2 where t1.a=t2.a;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	SIMPLE	t2	ALL	NULL	NULL	NULL	NULL	3	Using where
-1	SIMPLE	t1	ref	PRIMARY	PRIMARY	26	test.t2.a	1	Using index condition(BKA); Using join buffer (flat, BKA join); Key-ordered scan
+1	SIMPLE	t1	ref	PRIMARY	PRIMARY	26	test.t2.a	1	Using where; Using join buffer (flat, BKA join); Key-ordered scan
 select * from t1, t2 where t1.a=t2.a;
 a	b	filler	a	b
 a-1010=A	1010	filler	a-1010=A	1010
@@ -133,7 +133,7 @@ set join_cache_level=6;
 explain select * from t1, t2 where t1.a=t2.a and t2.b + t1.b > 100;
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 1	SIMPLE	t2	ALL	NULL	NULL	NULL	NULL	3	Using where
-1	SIMPLE	t1	ref	PRIMARY	PRIMARY	4	test.t2.a	1	Using index condition(BKA); Using join buffer (flat, BKA join); Key-ordered scan
+1	SIMPLE	t1	ref	PRIMARY	PRIMARY	4	test.t2.a	1	Using where; Using join buffer (flat, BKA join); Key-ordered scan
 select * from t1, t2 where t1.a=t2.a and t2.b + t1.b > 100;
 a	b	c	filler	a	b
 set optimizer_switch='index_condition_pushdown=off';