ClickHouse列式儲存數(shù)據(jù)庫
ClickHouse是俄羅斯第一大搜索引擎Yandex開發(fā)的列式儲存數(shù)據(jù)庫.令人驚喜的是,這個列式儲存數(shù)據(jù)庫的性能大幅超越了很多商業(yè)MPP數(shù)據(jù)庫軟件,比如Vertica,InfiniDB.
相比傳統(tǒng)的數(shù)據(jù)庫軟件,ClickHouse要快100-1000X:
100Million 數(shù)據(jù)集:
-
ClickHouse比Vertica約快5倍,比Hive快279倍,比My SQL快801倍
1Billion 數(shù)據(jù)集:
-
ClickHouse比Vertica約快5倍,MySQL和Hive已經(jīng)無法完成任務(wù)了
該項目當(dāng)前還有一些不足:
-
pre-build包只有Ubuntu平臺的可用,并且該項目當(dāng)前沒有任何架構(gòu)文檔
-
只有Github上面的C++源代碼
主要功能
-
True column-oriented
-
Vectorized query execution
-
Data compression
-
Parallel and distributed query execution
-
Real-time data ingestion
-
On-disk locality of reference
-
Real-time query processing
-
Cross-datacenter replication
-
High availability
-
SQL support
-
Local and distributed joins
-
Pluggable external dimension tables
-
Arrays and nested data types
-
Approximate query processing
-
Probabilistic data structures
-
Full support of IPv6
-
Features for web analytics
-
State-of-the-art algorithms
-
Detailed documentation
-
Clean documented code
應(yīng)用場景
-
Web and App analytics
-
Advertising networks and RTB
-
Telecommunications
-
E-commerce
-
Information security
-
Monitoring and telemetry
-
Business intelligence
-
Online games
-
Internet of Things
