- About Uqbar
- Release Note
- Uqbar Installation
- Uqbar Management
- Data Retention Policy
- Time-Series Table Management
- Time-Series Data Write
- Data Compression
- Data Deletion
- Data Query
- Continuous Aggregation
- Time-Series Views
- Cluster Management
- Backup and Restoration
- Security
- GUC Parameters
- SQL Syntax
- Third Party Tools Support
- Glossary
Uqbar Hyper-Converged Time-Series Database
Introduction
Uqbar is a hyper-converged time-series database developed based on the openGauss kernel. With support for both relational and time-series models, it is specially designed for IoT (Internet of Things) scenarios, featuring high performance, low cost, stability and reliability, and openness and ease of use. Uqbar is able to manage massive time-series data, and manage relational data in OLTP scenarios. It also supports complex association query across time-series and relational data, providing a one-stop data solution for IoT.
Customer Benefits
Hyper-Converged
Uqbar offers one-stop hyper-converged data processing capabilities, meeting multi-model data management requirements in IoT scenarios.
-
Support for both relational and time-series models: In IoT scenarios, there is a need to manage sensor metadata and sensor monitoring data. The traditional solution needs to use a set of database system to manage each, which increases the purchase and maintenance costs. With the Uqbar hyper-converged time-series database, you can manage sensor metadata and monitoring time-series data in one database at the same time.
-
Support for complex association query across time-series and relational data: One SQL statement can help complete complex association query across relational and time-series data. It supports rich time-series data analysis and query requirements, and can also effectively reduce the business complexity of the upper analysis platform, thereby improving the query and analysis efficiency.
Extreme Performance
Uqbar provides industry-leading performance and cost capabilities to meet the massive data management requirements in IoT scenarios.
- Constant write of massive sensor data under great workload: With technical optimization achieved in terms of NUMA, intelligent data partitioning, reuse of execution plans, and rapid expansion of data pages, the writing performance of a single node in high concurrency scenarios can reach up to 8 million points per second, exceeding the industry benchmark by 20%.
- High-performance aggregated query and analysis of massive data: With such technologies as column-store engine, vectorized query engine, and concurrent query, the aggregated query and analysis performance is optimized to make the response time reach the millisecond level, exceeding the industry benchmark by 30%.
- Multi-dimensional combination query: With the inverted index capability, multi-dimensional arbitrary combination query is supported. As a result the multi-dimensional query performance is 3 times more than that of other products of the same kind.
- Extreme data compression ratio: With the column-store and time-series compression algorithm, the data compression ratio is increased in 10 folds in typical scenarios while the storage cost is reduced and the query efficiency is improved.
- Management of massive sensor indicators: A single node supports management of tens of millions of sensor indicators.
Secure and Reliable
Uqbar provides perfect high availability (HA) and security protection capabilities to meet the security, reliability, and compliance requirements of the IoT scenario.
-
Primary/standby HA: provides the primary/standby cluster capability and has a built-in HA component. It supports automatic fault detection and failover, ensuring service continuity.
-
Data consistency: provides the ACID guarantee for relational data and allows users to customize the consistency level for time-series data.
-
Perfect data security and protection system: supports fine-grained permission management, communication encryption, operation audit, backup and recovery, and other functions, and provides all-round security protection capabilities before, in, and after the event.
Open and Easy-to-Use
Uqbar provides a wealth of ecological connection and O&M management tools to meet the simple and easy-to-use requirements of the IoT scenario.
-
Multiple data write methods: provides client drivers in C/C++, Java, Go and other languages to simplify application development. It also supports data consumption from Kafka.
-
Standard SQL: supports data write and query using standard SQLs, almost zero learning cost
-
Data visualization: supports seamless integration with Grafana and facilitates visualization and insight of time-series data through Grafana.
-
Graphical management tool: provides a graphical database O&M management tool to support automatic installation and deployment, database monitoring, backup and recovery management, alarm management and other functions, thereby improving the database O&M management efficiency.
Application Scenarios
IoT
With the hyper-convergence, extreme performance, security, reliability, and openness and ease of use capabilities, Uqbar can meet the needs of IoT scenarios for low-cost storage, high-performance writing, complex analysis and query of massive sensor monitoring data, and continuous high availability of the system.
Intelligent O&M
The powerful query capability of Uqbar includes multi-dimensional arbitrary combination queries, association analysis queries across relational table and time-series table, and extreme query performance, which can meet the complex analysis and query requirements of massive data.