ApsaraDB for ClickHouse is an Alibaba Cloud managed service for ClickHouse. ClickHouse is an open source column-oriented database management system (DBMS). Compared with open source ClickHouse, ApsaraDB for ClickHouse is more stable, secure, and maintainable. ApsaraDB for ClickHouse allows you to purchase cloud resources and create ApsaraDB for ClickHouse clusters on Alibaba Cloud with ease.
Introduction video for ApsaraDB for ClickHouse
Features
ApsaraDB for ClickHouse is a column-oriented DBMS for online analytical processing (OLAP). ApsaraDB for ClickHouse supports SQL queries and provides good query performance. In particular, ApsaraDB for ClickHouse provides an order of magnitude faster aggregate analysis and queries on wide tables than other analytical databases. ApsaraDB for ClickHouse provides the following features:
High data compression ratio
Multi-core parallel computing
Compute engine for vectorization
Support for nested data structures
Support for sparse indexes
For more information about ApsaraDB for ClickHouse, see the official website of ClickHouse and ClickHouse documentation.
Architecture
Compared with open source ClickHouse, ApsaraDB for ClickHouse is more stable, secure, and maintainable. The architecture of ApsaraDB for ClickHouse is similar to that of open source ClickHouse, as shown in the following figure.
ApsaraDB for ClickHouse clusters use an architecture in which computing is coupled with storage. ApsaraDB for ClickHouse clusters have the following features:
All nodes are peer nodes. Each node can compute and manage background data, and process query and write requests.
Each ApsaraDB for ClickHouse cluster contains one or more shards. Each shard contains one or more replicas.
All nodes are deployed on Alibaba Cloud Elastic Compute Service (ECS) instances. At the underlying layer, highly reliable disks are used as the persistent storage.
Enhanced features
Enhanced security: provides features such as user management and whitelists.
Graphical O&M: provides features such as the version update and failover.
Scalability: supports a variety of specifications and elastic scaling.
Data ecology: supports convenient data synchronization and migration with other data sources in the cloud.