The Search service is a solution for complex multidimensional queries and full-text search.
The full-text index feature is disabled for ApsaraDB for HBase Performance-enhanced Edition. Its capabilities have been fully migrated to Lindorm and enhanced to be more powerful, mature, and user-friendly.
Solr is an enterprise-grade search platform that is built on Apache Lucene and is one of the top platforms for distributed full-text search. Solr supports various complex conditional queries and full-text search. Solr is used by a wide range of customers. Deeply integrated with Apache Solr, ApsaraDB for HBase provides the Search service that allows you to store large amounts of data and supports complex multidimensional queries and full-text search.
The Search service is suitable for business scenarios in which you need to store large amounts of data and query data based on a combination of conditions. Example:
In logistics scenarios, you need to store massive tracking data and query specific information based on a combination of multiple fields.
In traffic monitoring scenarios, you need to store a large number of traffic records and retrieve the required information based on a combination of vehicle information.
In website member and product information retrieval scenarios, you need to store a large amount of product and membership information, and support complex queries based on conditions to meet the needs of website users.
The preceding figure shows the overall data flow of the Search service. After data is written to ApsaraDB for HBase Performance-enhanced Edition, Lindorm Tunnel Service (LTS, formerly known as BDS) synchronizes the data to Apache Solr in real time. The ApsaraDB for HBase cluster, LTS server, and Apache Solr cluster are independent of each other. This architecture allows you to manage each cluster separately. For example, if the processing capability of the Apache Solr cluster is insufficient, you can scale out the Solr cluster only. If the synchronization capability of the LTS server is insufficient, you can scale out the LTS server individually. This architecture allows you to select applicable servers of ApsaraDB for HBase, LTS, and Apache Solr to fit different scenarios. In addition, this architecture significantly improves the stability of the system.
Comparison with secondary indexes
ApsaraDB for HBase Performance-enhanced Edition provides Secondary indexes, which allows you to query data based on non-primary key columns at a low cost. ApsaraDB for HBase Performance-enhanced Edition is suitable for scenarios in which you need to query tables that contain only a few columns. If complex multidimensional queries are required in your business, we recommend that you use the Search service.
Comparison with Apache Solr
The Search service is developed based on the deep integration of ApsaraDB for HBase Performance-enhanced Edition and Apache Solr. You can use HBase Shell or perform API operations to associate the ApsaraDB for HBase cluster with the Apache Solr cluster, without the need to monitor the status of each service.
The Search service is designed based on Apache Solr and is fully compatible with Solr APIs. The Search service provides a reliable and high-performance enterprise-grade search platform with improved system stability, high read/write performance, and monitoring and alert features.
Activate the Search service
To activate the Search service, perform the following steps:
Create an ApsaraDB for HBase Performance-enhanced Edition cluster.
Create an LTS cluster.
After the ApsaraDB for HBase Performance-enhanced Edition cluster is created, click
Full-text Indexin the ApsaraDB for HBase console to purchase and associate the Search instance with the ApsaraDB for HBase Performance-enhanced Edition cluster.
For more information, see Activate the Search service.
User Guide
For more information, see Quick start and Manage indexes.
Best practices
For more information, see Update the configuration set.