Commit Graph

14 Commits

Author SHA1 Message Date
Ivan Babrou
d9d8d0242e Optimize PK lookup for queries that match exact PK range
Existing code that looks up marks that match the query has a pathological
case, when most of the part does in fact match the query.

The code works by recursively splitting a part into ranges and then discarding
the ranges that definitely do not match the query, based on primary key.

The problem is that it requires visiting every mark that matches the query,
making the complexity of this sort of look up O(n).

For queries that match exact range on the primary key, we can find
both left and right parts of the range with O(log 2) complexity.

This change implements exactly that.

To engage this optimization, the query must:

* Have a prefix list of the primary key.
* Have only range or single set element constraints for columns.
* Have only AND as a boolean operator.

Consider a table with `(service, timestamp)` as the primary key.

The following conditions will be optimized:

* `service = 'foo'`
* `service = 'foo' and timestamp >= now() - 3600`
* `service in ('foo')`
* `service in ('foo') and timestamp >= now() - 3600 and timestamp <= now`

The following will fall back to previous lookup algorithm:

* `timestamp >= now() - 3600`
* `service in ('foo', 'bar') and timestamp >= now() - 3600`
* `service = 'foo'`

Note that the optimization won't engage when PK has a range expression
followed by a point expression, since in that case the range is not continuous.

Trace query logging provides the following messages types of messages,
each representing a different kind of PK usage for a part:

```
Used optimized inclusion search over index for part 20200711_5710108_5710108_0 with 9 steps
Used generic exclusion search over index for part 20200711_5710118_5710228_5 with 1495 steps
Not using index on part 20200710_5710473_5710473_0
```

Number of steps translates to computational complexity.

Here's a comparison for before and after for a query over 24h of data:

```
Read 4562944 rows, 148.05 MiB in 45.19249672 sec.,   100966 rows/sec.,   3.28 MiB/sec.
Read 4183040 rows, 135.78 MiB in 0.196279627 sec., 21311636 rows/sec., 691.75 MiB/sec.
```

This is especially useful for queries that read data in order
and terminate early to return "last X things" matching a query.

See #11564 for more thoughts on this.
2020-07-11 12:26:54 -07:00
Alexey Milovidov
8872417d00 Respect direct_io/mmap settings while reading secondary indices 2020-06-25 22:31:54 +03:00
alesapin
b1e8976df4 Merge with master 2020-06-22 12:04:27 +03:00
alesapin
dffdece350 getColumns in StorageInMemoryMetadta (only compilable) 2020-06-17 19:39:58 +03:00
alesapin
1afdebeebd Primary key in storage metadata 2020-06-17 15:39:20 +03:00
alesapin
71f99a274d Compileable getSampleBlockWithColumns in StorageInMemoryMetadata 2020-06-16 17:25:08 +03:00
alesapin
3847ea892d Merge branch 'master' into consistent_metadata3 2020-06-01 13:17:59 +03:00
Alexey Milovidov
25f941020b Remove namespace pollution 2020-05-31 00:57:37 +03:00
alesapin
5f8b69547b More readable code 2020-05-28 16:45:08 +03:00
Nikolai Kochetov
4d22374f24 Merged with master. 2020-05-14 12:06:15 +03:00
Anton Popov
84d4ad4315 add some comments 2020-05-13 19:00:06 +03:00
Anton Popov
67213b8ad4 fix sample with final 2020-05-12 21:23:40 +03:00
Nikolai Kochetov
19dadb8c2d Add parallel final. 2020-04-30 12:59:08 +03:00
Ivan Lezhankin
06446b4f08 dbms/ → src/ 2020-04-03 18:14:31 +03:00