High-Performance Multidimensional Analysis and Query in Real Time
Hologres is a cloud-native Hybrid Serving & Analytical Processing (HSAP) system that is seamlessly integrated with the big data ecosystem. You can use Hologres to analyze and process PB-scale data with high concurrency and low latency. With Hologres, you can use your business intelligence (BI) tools to analyze data in multiple dimensions and explore your business in real time.
Hologres eliminates the disadvantages of traditional real-time data warehousing solutions, such as data silos and data redundancy. It meets your requirements for data migration and real-time analysis of large volumes of data.
Responds to queries on PB-scale data at sub-second speeds. This response speed allows you to analyze data in multiple dimensions and explore your business in real time.
Supports highly concurrent writes and queries at a speed of up to 100 million transactions per second (TPS). Data can be queried immediately after it is written.
Uses the compute-storage separation architecture. It allows you to dynamically upgrade or downgrade configurations and resize the storage capacity based on your business requirements.
Reads data from offline data warehouses of MaxCompute for analysis.
Stores multiple replicas of data by default based on the distributed high-availability architecture. The SLA level reaches 99.99%.
Support for Various Query Modes
Stores data in column- or row-oriented mode. It also allows you to conduct simple, complex, or ad-hoc queries. Hologres can process a large volume of SQL statements concurrently in distributed mode. It returns query responses on large volumes of data at sub-second speeds. The resource utilization is high.
Complex Query on Big Data
Supports parallel computing and responds to federated analytics of PB-scale data at sub-second speeds.
Point Query on Big Data
Stores PB-scale data and writes or queries 100 million records per second. This is more than 10 times the performance of open source components.
Seamlessly connects to MaxCompute. You can use Hologres to perform interactive analysis without the need to migrate data. Query results are quickly generated. You can use Hologres to query MaxCompute data or implement federated computing based on MaxCompute data and real-time data.
High Compatibility and Ease of Development
Compatible with the PostgreSQL ecosystem and seamlessly integrated with DataWorks. DataWorks is the big data computing engine and big data development platform of Alibaba Cloud. You can get started with Hologres without additional learning.
Compatibility with PostgreSQL
Compatible with the PostgreSQL ecosystem. Hologres provides the JDBC or ODBC API to interconnect with third-party extract, transform, and load (ETL) and BI tools, such as Tableau, QuickBI, and FineBI.
Integration with DataWorks
Seamlessly integrated with DataWorks to provide graphic, intelligent, and unified data warehousing and interactive analysis.
Integration with Proxima
Seamlessly connects to PAI and uses Proxima as a built-in vector search plug-in to support real-time computing of feature data and real-time vector retrieval. Proxima is a vector search engine developed by Alibaba DAMO Academy.
Writes and Updates Data in Real Time
Supports highly-concurrent writes and updates at a speed of up to 100 million TPS. Data can be queried immediately after it is written.
Best Practices of Real-time Data Warehousing Based on Integration with Realtime Compute for Apache Flink
Integrated with Cloud Native Realtime Compute for Apache Flink to support real-time, high-throughput data writes and updates. Built-in connectors are used to support different types of tables, such as source tables, result tables, and dimension tables.
Log on to the Hologres console and follow the purchase wizard to create a Hologres instance.
Connect to Development Tool
Connect to a terminal tool, open source tool, or Alibaba Cloud tool based on your data development requirements.
Hologres is compatible with PostgreSQL 11. Use standard PostgreSQL statements to grant permissions.
You can use standard PostgreSQL statements to perform interactive analysis and development or use open source or commercial management tools to design and manage databases.
Pay-as-you-go applies to scenarios when computing or storage resources are unspecified.
You are billed based on the running durations of your instances, the size of stored data, and storage durations.
Subscription applies to scenarios when Hologres is used for real-time analysis in the long term.
You need to estimate the required computing and storage resources based on your actual business scenario and purchase this service in subscription mode.