Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Not a Bug
-
10.5(EOL)
-
None
Description
r_other_time_ms was introduced to collect the time spent in JOIN "between the table accesses" - checking the WHERE, etc.
However it is also printed for tables that are put as a source to filesort:
analyze format=json select char_field, count(distinct b) from t_char_1000000 group by char_field;
|
| {
|
"query_block": {
|
"select_id": 1,
|
"r_loops": 1,
|
"r_total_time_ms": 4048.4,
|
"read_sorted_file": {
|
"r_rows": 1e6,
|
"filesort": {
|
"sort_key": "t_char_1000000.char_field",
|
"r_loops": 1,
|
"r_total_time_ms": 1634.5,
|
"r_used_priority_queue": false,
|
"r_output_rows": 1000000,
|
"r_sort_passes": 1,
|
"r_buffer_size": "2047Kb",
|
"r_sort_mode": "packed_sort_key,rowid",
|
"table": {
|
"table_name": "t_char_1000000",
|
"access_type": "ALL",
|
"r_loops": 1,
|
"rows": 1000000,
|
"r_rows": 1e6,
|
"r_table_time_ms": 2267.4,
|
"r_other_time_ms": 1781,
|
"filtered": 100,
|
"r_filtered": 100
|
}
|
}
|
}
|
}
|
Does it have any meaning there? Need to check this.