title | date | summary | tags | ||||
---|---|---|---|---|---|---|---|
Weekly update (January 15 ~ January 21, 2018) |
2018-01-22 |
Last week, we landed 43 PRs in the TiDB repositories, 13 PRs in the TiSpark repositories, and 18 PRs in the TiKV and PD repositories. |
|
Last week, we landed 43 PRs in the TiDB repositories.
- Optimize the
com_field_list
command and makeUse Database
faster - Improve the sort efficiency on
lookupTableTask.rows
- Enlarge the default
distsql_scan_concurrency
- Log a warning when the memory usage of
HashJoinExec
exceeds threshhold - Refine the behavior when updating
StatsDelta
- Split the presentation and evaluation layers of aggregation functions
- Use
BatchGet
to speed upLOAD DATA
Last week, we landed 13 PRs in the TiSpark repositories.
- Refactor the type and expression systems totally
- Reorganize the type and expressions
- Use the visitor instead of raw expressions when processing traversal
- Fixed
type infer
logics - Refactor predicates processing logic according to the visitor pattern
- Fixed index encoding
- Change
CodecDataInput
to use non-synchronizedByteArrayInputStream
- Add TiKV, PD, and TiDB's config files for Docker test environment
Last week, we landed 18 PRs in the TiKV and PD repositories.