[
https://issues.apache.org/jira/browse/IGNITE-10329?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16710010#comment-16710010
]
Pavel Kuznetsov commented on IGNITE-10329:
------------------------------------------
Results of the run
For select-range benchmarks:
||Vendor\Benchmark (Q/sec, Avg) ||SelectBySalary ||SelectByPk
||SelectBySalaryWithJoin ||SelectByPkWithJoin||
||ignite thin inmemory |885.36 |1,030.81 |170.41 |189.30|
||ignite jdbc v2 inmemory |891.48 |1,046.89 |157.34 |170.51|
||ignite thin persistence |560.58 |669.73 |141.23 |146.14|
||ignite jdbc v2 persistence |582.89 |653.87 |128.28 |137.35|
||Mysql (tunned) |770.97 |786.83 |199.29 |237.16|
||PgSQL (tunned) |1,845.70 |1,795.23 |73.94 |74.05|
Pgsql plan for SelectBySalaryWithJoin :
{noformat}
Hash Join (cost=1292.42..2624.51 rows=361 width=1572) Hash Cond:
(o.id = p.org_id) -> Seq Scan on organization o (cost=0.00..839.04
rows=10304 width=524)
-> Hash (cost=1287.91..1287.91 rows=361 width=1056)
-> Bitmap Heap Scan on person p (cost=60.12..1287.91 rows=361 width=1056)
Recheck Cond: ((salary >= '366018000'::bigint) AND
(salary <= '367017000'::bigint)) -> Bitmap Index
Scan on sal_idx (cost=0.00..60.03 rows=361 width=0)
Index Cond: ((salary >= '366018000'::bigint) AND (salary <=
'367017000'::bigint))
{noformat}
Pgsql plan for SelectByPkWithJoin :
{noformat}
Hash Join (cost=1292.42..2624.51 rows=361 width=1572) Hash Cond:
(o.id = p.org_id) -> Seq Scan on organization o (cost=0.00..839.04
rows=10304 width=524)
-> Hash (cost=1287.91..1287.91 rows=361 width=1056)
-> Bitmap Heap Scan on person p (cost=60.12..1287.91 rows=361 width=1056)
Recheck Cond: ((id >= '661420'::bigint) AND (id <=
'662419'::bigint)) -> Bitmap Index Scan on
person_pkey (cost=0.00..60.03 rows=361 width=0)
Index Cond: ((id >= '661420'::bigint) AND (id <= '662419'::bigint))
{noformat}
Results for benchmarks that fetch all database:
||Vendor\Benchmark (Q/sec, Avg) ||SelectAll ||SelectAllJoin||
||ignite thin inmemory |1.28 |0.20|
||ignite jdbc v2 inmemory |1.52 |0.19|
||ignite thin persistence |0.91 |0.16|
||ignite jdbc v2 persistence |1.02 |0.16|
||Mysql (tunned) |1.69 |0.31|
||PgSQL (tunned) |4.29 |0.90|
> Create JDBC "query" and "query join" benchmarks and compare them with
> Postgres and MySQL
> ----------------------------------------------------------------------------------------
>
> Key: IGNITE-10329
> URL: https://issues.apache.org/jira/browse/IGNITE-10329
> Project: Ignite
> Issue Type: Task
> Components: sql, yardstick
> Reporter: Vladimir Ozerov
> Assignee: Pavel Kuznetsov
> Priority: Major
> Fix For: 2.8
>
>
> Currently we have {{IgniteSqlQueryBenchmark}} and
> {{IgniteSqlQueryJoinBenchmark}} benchmarks which query data over salary range
> and optionally joins it with second table. Let's create a set of similar
> benchmarks which will use JDBC to load and query data, and execute them
> against one-node Ignite cluster, MySQL and Postgres.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
