[CORE-389] Totalnum performance improvements Created: 12/Mar/20 Updated: 30/May/22 Resolved: 27/Apr/22 |
|
| Status: | Closed |
| Project: | i2b2 Core Software |
| Component/s: | None |
| Affects Version/s: | 1.7.12 |
| Fix Version/s: | 1.7.13, 1.7.12a |
| Type: | Improvement | Priority: | Major |
| Reporter: | Jeffrey Klann | Assignee: | Jeffrey Klann |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Σ Remaining Estimate: | Not Specified | Remaining Estimate: | Not Specified |
| Σ Time Spent: | Not Specified | Time Spent: | Not Specified |
| Σ Original Estimate: | Not Specified | Original Estimate: | Not Specified |
| Sub-Tasks: |
|
||||||||||||||||||||
| Rank: | 0|s000e0: | ||||||||||||||||||||
| Participant/s: |
| Description |
|
Totalnum counter performance is unacceptable on very large (>1.5million) element ontologies such as the ACT Medications ontology. Even smaller but very deep ontologies fail to compute on MSSQL and Oracle. Also, we plan to add some additional debug messages.
|
| Comments |
| Comment by Jeffrey Klann [ 24/Apr/20 ] |
| 1.7.12a has performance improvements in Postgres and MSSQL. Oracle improvements still forthcoming. |
| Comment by Jeffrey Klann [ 27/Apr/22 ] |
|
Oracle, PGSQL, and MSSQL are all now on par with performance. MSSQL's most recent improvement, contributed by Darren Henderson, does not make sense in the architecture of the other scripts (it doesn't generate the same temp table).
This could be reopened in the future, as performance can always be improved. |
| Comment by Reeta Metta [ 30/May/22 ] |
| oracle script errors out on |