Details
-
Bug
-
Status: Closed (View Workflow)
-
Blocker
-
Resolution: Fixed
-
10.6
Description
The fix of MDEV-24671 introduced a serious performance regression that is observable at 32 concurrent connections.
My current hypothesis based on some initial investigation is that the changed sizeof(trx->lock) caused an increase of cache misses.
Attachments
Issue Links
- causes
-
-
- Closed
-
-
-
- Closed
-
-
-
- Closed
-
- is caused by
-
-
- Closed
-
- relates to
-
-
- Closed
-
Activity
| Field | Original Value | New Value |
|---|---|---|
| Link |
This issue is caused by |
| Status | Open [ 1 ] | In Progress [ 3 ] |
| issue.field.resolutiondate | 2021-02-05 17:15:44.0 | 2021-02-05 17:15:44.751 |
| Fix Version/s | 10.6.0 [ 24431 ] | |
| Fix Version/s | 10.6 [ 24028 ] | |
| Resolution | Fixed [ 1 ] | |
| Status | In Progress [ 3 ] | Closed [ 6 ] |
Some performance regression is still present. The function lock_wait() that was introduced in MDEV-24671 was holding lock_sys.wait_mutex for unnecessarily long time.
| Resolution | Fixed [ 1 ] | |
| Status | Closed [ 6 ] | Stalled [ 10000 ] |
| Status | Stalled [ 10000 ] | In Progress [ 3 ] |
| Fix Version/s | 10.6 [ 24028 ] | |
| Fix Version/s | 10.6.0 [ 24431 ] |
Even after reducing the lock_sys.wait_mutex hold time in lock_wait() to the minimum, some regression is present, and more work needs to be done.
| Link |
This issue relates to |
It looks like MDEV-24671 emphasized a pre-existing bottleneck on log_sys.mutex, which we hope to address in MDEV-14425. The remaining performance regression was only observed on RAM disk, not on real storage.
That said, the latest change that is being tested should reduce contention on lock_sys.latch and lock_sys.wait_mutex to the absolute minimum.
| issue.field.resolutiondate | 2021-03-02 12:41:36.0 | 2021-03-02 12:41:36.866 |
| Fix Version/s | 10.6.0 [ 24431 ] | |
| Fix Version/s | 10.6 [ 24028 ] | |
| Resolution | Fixed [ 1 ] | |
| Status | In Progress [ 3 ] | Closed [ 6 ] |
| Link |
This issue causes |
| Link |
This issue causes |
| Link |
This issue causes |
| Workflow | MariaDB v3 [ 118851 ] | MariaDB v4 [ 158862 ] |
Reducing sizeof(trx_lock_t) did not help (I did it anyway), but some refactoring to reduce the hold time of lock_sys.mutex and lock_sys.wait_mutex seems to have done the trick. Thanks to axel for helping with the analysis and validation!