[MDEV-738] LP:686353 - 2.7x performance regression with correlated indexes and sort_intersection in maria-5.1-wl21 Created: 2010-12-07 Updated: 2015-04-16 Resolved: 2015-04-16 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | Optimizer |
| Affects Version/s: | None |
| Fix Version/s: | N/A |
| Type: | Bug | Priority: | Minor |
| Reporter: | Philip Stoev (Inactive) | Assignee: | Igor Babaev |
| Resolution: | Won't Fix | Votes: | 0 |
| Labels: | Launchpad | ||
| Attachments: |
|
| Description |
|
The following query against the DBT-3 dataset
is 10x times slower when executed with sort_intersection. Igor says that this is " not interesting because lineitem is naturally ordered by l_receiptDATE and l_receiptDATE is strongly correlated with l_shipdate" , " in the case when indexes are correlated intersection causes degradation: this is a well know fact.". At the same time, the data set is supposedly standard and in real life, almost all date-based indexes will be naturally ordered and correlated to the PK. |
| Comments |
| Comment by Philip Stoev (Inactive) [ 2010-12-07 ] |
|
Re: 10x performance regression with correlated indexes and sort_intersection in maria-5.1-wl21 id: 1 4. Platform. 5. Execution time in maria-5.1-wl21 - 0.27 sec , Handler_read_next 72166 , Handler_read_rnd 6296 |
| Comment by Philip Stoev (Inactive) [ 2010-12-07 ] |
|
Re: 10x performance regression with correlated indexes and sort_intersection in maria-5.1-wl21 |
| Comment by Arjen Lentz (Inactive) [ 2010-12-07 ] |
|
Re: [Bug 686353] [NEW] 10x performance regression with correlated indexes and sort_intersection in maria-5.1-wl21 On 07/12/2010, at 5:45 PM, Philip Stoev wrote: I don't care very much for the specific scenario, I think the main Regards, Follow us at http://openquery.com/blog/ & http://twitter.com/openquery |
| Comment by Philip Stoev (Inactive) [ 2010-12-13 ] |
|
Re: 10x performance regression with correlated indexes and sort_intersection in maria-5.1-wl21 |
| Comment by Philip Stoev (Inactive) [ 2010-12-13 ] |
|
dbt3 dataset scale 0.1 |
| Comment by Rasmus Johansson (Inactive) [ 2011-01-26 ] |
|
Launchpad bug id: 686353 |
| Comment by Elena Stepanova [ 2015-04-16 ] |
|
The optimization is off by default, and as the description says the scenario wasn't considered interesting to begin with, so I suppose it's safe to assume it won't be fixed. Closing as such. |