Industry Algorithm Edition
Intelligent | Flexible and customizable | Secure and stable |
Offers built-in, customizable algorithm models and industry-specific retrieval and ranking algorithms tailored to the search patterns of different industries. | Customize algorithm models, application structures, data processing, query analysis, and ranking configurations to meet business requirements, improve click-through rates, and shorten development cycles. |
|
Auto scaling | Rich features | Out-of-the-box |
Supports auto scaling to adjust resources based on demand. | Supports additional search features including top searches, hints, drop-down suggestions, and report statistics to visualize and analyze search data. | Provides a fully managed search service, eliminating the need to deploy or manage a cluster. |
High-performance Search Edition
High throughput | Secure and stable | Auto scaling | Out-of-the-box |
A single table can handle tens of thousands of write transactions per second (TPS), and updates take effect within seconds. |
| Supports auto scaling to adjust resources based on demand. | Provides a fully managed search service, eliminating the need to deploy or manage a cluster. |
Vector Search Edition
Stable | Efficient | Cost-effective |
The underlying C++ engine has been production-hardened for over a decade and powers multiple core business services, making it well suited for mission-critical search applications. | As a distributed search engine, it handles large-scale data retrieval and real-time updates that take effect in seconds, suiting applications that require low query latency and high data timeliness. | Supports multiple index compression strategies and multi-value index load tests, to meet query requirements at a lower cost. |
Vector algorithms | SQL query | |
Supports vector retrieval for various types of unstructured data, including voice, images, video, natural language, and behavioral data. | Supports SQL query syntax and online multi-table joins. It offers a library of built-in user-defined functions (UDFs) and allows UDF customization to meet diverse retrieval needs. The O&M system includes an integrated SQL Studio for SQL development and testing. | |
Retrieval Engine Edition
Stable | Efficient | Cost-effective |
The underlying C++ engine has been production-hardened for over a decade and powers multiple core business services, making it well suited for mission-critical search applications. | As a distributed search engine, it handles large-scale data retrieval and real-time updates that take effect in seconds, suiting applications that require low query latency and high data timeliness. | Supports multiple index compression strategies and multi-value index load tests, to meet query requirements at a lower cost. |
Feature-rich | SQL query | |
Supports multiple analyzer types, various index types, and a query syntax to meet a wide range of retrieval needs. It also provides a plugin mechanism to customize business logic. | Supports SQL query syntax and online multi-table joins. It offers a library of built-in user-defined functions (UDFs) and allows UDF customization to meet diverse retrieval needs. The O&M system includes an integrated SQL Studio for SQL development and testing. | |