Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) is an online query acceleration service for Hologres scenarios. Based on a Hologres cloud-native architecture that separates storage from computing, this service uses shared cluster resources to accelerate queries on data in MaxCompute tables. This topic describes the relevant content about Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition).

Scenarios

You can use Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) in the following scenarios:
  • Scenarios in which accelerated queries on data in MaxCompute tables are required.
  • Scenarios in which data in MaxCompute tables is infrequently queried and the query latency must be low.
  • Scenarios in which a large amount of data in MaxCompute tables is queried by using complex queries and dedicated Hologres instances with low specifications cannot perform the complex queries.
  • Scenarios in which queries on data in MaxCompute tables are accelerated and business intelligence (BI) tools are connected to Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) for data analysis. For example, you can connect Tableau, Quick BI, and FineBI to Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition).

Benefits

Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) has the following benefits:
  • Provides PostgreSQL-compatible query API operations for seamless connection to a MaxCompute data store, without the need to migrate data. This way, a reliable, rapid, and cost-effective query service is provided for data analysis to query data within seconds.
  • Uses a serverless architecture to eliminate limits on resource scalability and elasticity. This way, resources are elastic and dynamically scalable and the O&M cost is saved.
  • Compared with MaxCompute, Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) improves the query speed on data in MaxCompute tables by 5 to 10 times.

Features

Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) instances share resources. As an individual object, each instance also has the following specific features:
  • Built-in query engine
    • Allows you to create foreign tables to accelerate queries on data in MaxCompute tables. When you query data in this way, the data is still stored in the MaxCompute tables.
    • Does not allow you to create internal tables or import data to Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) for query. If you need to import data to a Hologres internal table for query, you can purchase a dedicated Hologres instance.
  • Standard SQL statements

    Compatible with PostgreSQL 11, Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) provides standard PostgreSQL statements to query data.

  • Permission management

    You can manage permissions separately for each Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) instance in the same way as you manage permissions for dedicated Hologres instances. You can add or remove users of the current instance, or grant different access permissions to different users based on your business requirements.

Note Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) instances support all features of dedicated Hologres instances except for features related to internal tables and the data import feature.

Comparison of service types

The following table compares the features of a dedicated Hologres instance, Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition), and Lightning.
Item Dedicated Hologres instance Hologres Shared Cluster (BI-based MaxCompute Acceleration Edition) Lightning
Feature Category PostgreSQL 11 is supported. PostgreSQL 11 is supported. PostgreSQL 8.2 is supported.
Parameter optimization Parameter optimization Supported. For example, you can adjust parameters based on databases or users. Supported. For example, you can adjust parameters based on databases or users. Custom parameter optimization is not supported. For example, you cannot adjust parameters based on time zones.
Number of SQL connections or queries per second (QPS) The number varies based on the instance type. You can execute the show max_connections statement to query the number of SQL connections. The default number of SQL connections is 100. The fixed number of SQL connections is 20.
Data query based on foreign tables Meta module loading Data queries can be accelerated to respond within milliseconds. Data queries can be accelerated to respond within milliseconds. The query response is slow, but the response speed increases linearly with the number of tables in a MaxCompute project.
Data type The basic data types, DECIMAL type in MaxCompute V2.0, DATETIME type, and ARRAY type are supported. The basic data types, DECIMAL type in MaxCompute V2.0, DATETIME type, and ARRAY type are supported. Only basic data types are supported.
Engine optimization N/A The query performance is much better than that of Lightning. N/A
Data query based on internal tables Supported indexes Bitmap and clustered indexes are supported. None None
Specified distribution keys The distribution keys are specified to perform the Local Join or Local Group By operation. None None
Resources Computing and storage resources that you have purchased are exclusive to you. Computing resources are shared and storage resources are exclusive to you. Computing resources are shared and storage resources are exclusive to you.