thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-14 03:25:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #45444 from ClickHouse/index-docs

Browse Source 
Docs: Beautify section on secondary index types
This commit is contained in:
Robert Schulze 2023-01-19 19:36:45 +01:00 committed by GitHub
commit c7d12564cb
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 21 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
docs/en/engines/table-engines/mergetree-family/mergetree.md
View File

@ -390,40 +390,46 @@ SELECT count() FROM table WHERE u64 * i32 == 10 AND u64 * length(s) >= 1234
### Available Types of Indices {#available-types-of-indices}
#### `minmax`
#### MinMax
Stores extremes of the specified expression (if the expression is `tuple`, then it stores extremes for each element of `tuple`), uses stored info for skipping blocks of data like the primary key.
#### `set(max_rows)`
Syntax: `minmax`
#### Set
Stores unique values of the specified expression (no more than `max_rows` rows, `max_rows=0` means “no limits”). Uses the values to check if the `WHERE` expression is not satisfiable on a block of data.
#### `ngrambf_v1(n, size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
Syntax: `set(max_rows)`
Stores a [Bloom filter](https://en.wikipedia.org/wiki/Bloom_filter) that contains all ngrams from a block of data. Works only with datatypes: [String](/docs/en/sql-reference/data-types/string.md), [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) and [Map](/docs/en/sql-reference/data-types/map.md). Can be used for optimization of `EQUALS`, `LIKE` and `IN` expressions.
#### Bloom Filter
Stores a [Bloom filter](https://en.wikipedia.org/wiki/Bloom_filter) for the specified columns. An optional `false_positive` parameter with possible values between 0 and 1 specifies the probability of receiving a false positive response from the filter. Default value: 0.025. Supported data types: `Int*`, `UInt*`, `Float*`, `Enum`, `Date`, `DateTime`, `String`, `FixedString`, `Array`, `LowCardinality`, `Nullable`, `UUID` and `Map`. For the `Map` data type, the client can specify if the index should be created for keys or values using [mapKeys](/docs/en/sql-reference/functions/tuple-map-functions.md/#mapkeys) or [mapValues](/docs/en/sql-reference/functions/tuple-map-functions.md/#mapvalues) function.
Syntax: `bloom_filter([false_positive])`
#### N-gram Bloom Filter
Stores a [Bloom filter](https://en.wikipedia.org/wiki/Bloom_filter) that contains all n-grams from a block of data. Only works with datatypes: [String](/docs/en/sql-reference/data-types/string.md), [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) and [Map](/docs/en/sql-reference/data-types/map.md). Can be used for optimization of `EQUALS`, `LIKE` and `IN` expressions.
Syntax: `ngrambf_v1(n, size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
- `n` — ngram size,
- `size_of_bloom_filter_in_bytes` — Bloom filter size in bytes (you can use large values here, for example, 256 or 512, because it can be compressed well).
- `number_of_hash_functions` — The number of hash functions used in the Bloom filter.
- `random_seed` — The seed for Bloom filter hash functions.
#### `tokenbf_v1(size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
#### Token Bloom Filter
The same as `ngrambf_v1`, but stores tokens instead of ngrams. Tokens are sequences separated by non-alphanumeric characters.
#### `bloom_filter([false_positive])` — Stores a [Bloom filter](https://en.wikipedia.org/wiki/Bloom_filter) for the specified columns.
Syntax: `tokenbf_v1(size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
The optional `false_positive` parameter is the probability of receiving a false positive response from the filter. Possible values: (0, 1). Default value: 0.025.
#### Special-purpose
Supported data types: `Int*`, `UInt*`, `Float*`, `Enum`, `Date`, `DateTime`, `String`, `FixedString`, `Array`, `LowCardinality`, `Nullable`, `UUID`, `Map`.
- An experimental index to support approximate nearest neighbor (ANN) search. See [here](annindexes.md) for details.
For `Map` data type client can specify if index should be created for keys or values using [mapKeys](/docs/en/sql-reference/functions/tuple-map-functions.md/#mapkeys) or [mapValues](/docs/en/sql-reference/functions/tuple-map-functions.md/#mapvalues) function.
There are also special-purpose and experimental indexes to support approximate nearest neighbor (ANN) queries. See [here](annindexes.md) for details.
The following functions can use the filter: [equals](/docs/en/sql-reference/functions/comparison-functions.md), [notEquals](/docs/en/sql-reference/functions/comparison-functions.md), [in](/docs/en/sql-reference/functions/in-functions), [notIn](/docs/en/sql-reference/functions/in-functions), [has](/docs/en/sql-reference/functions/array-functions#hasarr-elem), [hasAny](/docs/en/sql-reference/functions/array-functions#hasany), [hasAll](/docs/en/sql-reference/functions/array-functions#hasall).
Example of index creation for `Map` data type
## Example of index creation for Map data type
```
INDEX map_key_index mapKeys(map_column) TYPE bloom_filter GRANULARITY 1
@ -484,9 +490,6 @@ For example:
:::
## Approximate Nearest Neighbor Search Indexes [experimental] {#table_engines-ANNIndex}
In addition to skip indices, there are also [Approximate Nearest Neighbor Search Indexes](/docs/en/engines/table-engines/mergetree-family/annindexes.md).
## Projections {#projections}
Projections are like [materialized views](/docs/en/sql-reference/statements/create/view.md/#materialized) but defined in part-level. It provides consistency guarantees along with automatic usage in queries.