thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-22 07:31:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

dbms: improved performance of aggregation by string key [#METR-2944].

Browse Source
This commit is contained in:
Alexey Milovidov 2014-05-10 06:42:45 +04:00
parent 82a44f8eec
commit 95827e2dc2
4 changed files with 74 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
dbms/include/DB/Common/HashTable/HashMap.h
View File

@ -74,6 +74,25 @@ struct HashMapCell
};
template <typename Key, typename TMapped, typename Hash, typename TState = HashTableNoState>
struct HashMapCellWithSavedHash : public HashMapCell<Key, TMapped, Hash, TState>
{
typedef HashMapCell<Key, TMapped, Hash, TState> Base;
size_t saved_hash;
HashMapCellWithSavedHash() : Base() {}
HashMapCellWithSavedHash(const Key & key_, const typename Base::State & state) : Base(key_, state) {}
HashMapCellWithSavedHash(const typename Base::value_type & value_, const typename Base::State & state) : Base(value_, state) {}
bool keyEquals(const Key & key_) const { return this->value.first == key_; }
bool keyEquals(const HashMapCellWithSavedHash & other) const { return saved_hash == other.saved_hash && this->value.first == other.value.first; }
void setHash(size_t hash_value) { saved_hash = hash_value; }
size_t getHash(const Hash & hash) const { return saved_hash; }
};
template
<
typename Key,
@ -112,3 +131,14 @@ template
typename Allocator = HashTableAllocator
>
using HashMap = HashMapTable<Key, HashMapCell<Key, Mapped, Hash>, Hash, Grower, Allocator>;
template
<
typename Key,
typename Mapped,
typename Hash = DefaultHash<Key>,
typename Grower = HashTableGrower<>,
typename Allocator = HashTableAllocator
>
using HashMapWithSavedHash = HashMapTable<Key, HashMapCellWithSavedHash<Key, Mapped, Hash>, Hash, Grower, Allocator>;

45
dbms/include/DB/Common/HashTable/HashSet.h
View File

@ -14,15 +14,16 @@
template
<
typename Key,
typename TCell,
typename Hash = DefaultHash<Key>,
typename Grower = HashTableGrower<>,
typename Allocator = HashTableAllocator
>
class HashSet : public HashTable<Key, HashTableCell<Key, Hash>, Hash, Grower, Allocator>
class HashSetTable : public HashTable<Key, TCell, Hash, Grower, Allocator>
{
public:
typedef HashSet<Key, Hash, Grower, Allocator> Self;
typedef HashTableCell<Key, Hash> Cell;
typedef HashSetTable<Key, TCell, Hash, Grower, Allocator> Self;
typedef TCell Cell;
void merge(const Self & rhs)
{
@ -55,3 +56,41 @@ public:
}
}
};
template <typename Key, typename Hash, typename TState = HashTableNoState>
struct HashSetCellWithSavedHash : public HashTableCell<Key, Hash, TState>
{
typedef HashTableCell<Key, Hash, TState> Base;
size_t saved_hash;
HashSetCellWithSavedHash() : Base() {}
HashSetCellWithSavedHash(const Key & key_, const typename Base::State & state) : Base(key_, state) {}
bool keyEquals(const Key & key_) const { return this->key == key_; }
bool keyEquals(const HashSetCellWithSavedHash & other) const { return saved_hash == other.saved_hash && this->key == other.key; }
void setHash(size_t hash_value) { saved_hash = hash_value; }
size_t getHash(const Hash & hash) const { return saved_hash; }
};
template
<
typename Key,
typename Hash = DefaultHash<Key>,
typename Grower = HashTableGrower<>,
typename Allocator = HashTableAllocator
>
using HashSet = HashSetTable<Key, HashTableCell<Key, Hash>, Hash, Grower, Allocator>;
template
<
typename Key,
typename Hash = DefaultHash<Key>,
typename Grower = HashTableGrower<>,
typename Allocator = HashTableAllocator
>
using HashSetWithSavedHash = HashSetTable<Key, HashSetCellWithSavedHash<Key, Hash>, Hash, Grower, Allocator>;

2
dbms/include/DB/Interpreters/Aggregator.h
View File

@ -48,7 +48,7 @@ typedef std::vector<AggregateDescription> AggregateDescriptions;
*/
typedef AggregateDataPtr AggregatedDataWithoutKey;
typedef HashMap<UInt64, AggregateDataPtr> AggregatedDataWithUInt64Key;
typedef HashMap<StringRef, AggregateDataPtr> AggregatedDataWithStringKey;
typedef HashMapWithSavedHash<StringRef, AggregateDataPtr> AggregatedDataWithStringKey;
typedef HashMap<UInt128, AggregateDataPtr, UInt128Hash> AggregatedDataWithKeys128;
typedef HashMap<UInt128, std::pair<StringRef*, AggregateDataPtr>, UInt128TrivialHash> AggregatedDataHashed;

2
dbms/include/DB/Interpreters/Set.h
View File

@ -94,7 +94,7 @@ private:
* одного или нескольких столбцов значений множеству.
*/
typedef HashSet<UInt64> SetUInt64;
typedef HashSet<StringRef> SetString;
typedef HashSetWithSavedHash<StringRef> SetString;
typedef HashSet<UInt128, UInt128Hash> SetHashed;
BlockInputStreamPtr source;