TSDB features highly efficient reading and writing of time series data. It allows you to write up to ten millions of data points per second, and the response time for reading one million data points is less than five seconds.
TSDB supports data writing through HTTP protocol and TSDB data writing protocol.
TSDB allows you to query data through HTTP APIs or the console. The console-based data query function supports data grouping, downsampling, space aggregation, and data visualization.
Data’s validity period settings
You can set the data’s validity period in the console or through APIs. After the data’s validity period is enabled and defined, the system marks the data as invalid once it expires, and automatically cleans it up at a specific time.
You can perform data cleanup by metric in the console or use APIs to do it more flexibly.
TSDB uses efficient data compression technology, which can reduce the average storage space required for a single data point to one to two bytes, so that the overall storage usage could be reduced by 90% and the data writing speed could be enhanced as well.
TSDB provides professional and comprehensive computing functions for time series data, and supports downsampling, data interpolation and space aggregation, which can meet the needs of various complicated data query scenarios.
TSDB provides an instance O&M system for you to monitor the instance running status, performance indicators, and storage space usage in real time, and set up the alarming mechanism to detect resource bottlenecks in time.
TSDB offers the following solutions to guarantee your data and instance security:
- Provide Virtual Private Cloud (VPC) network to fully ensure the security of the instance access.
- Provide the IP address whitelist function. You can set a whitelist of the hosts that are permitted to access your instance to further protect the access security of your instance and data. If a host is deployed within a VPC but not included in the whitelist, it cannot access your instance.
- In TSDB, the data is stored in three copies to fully assure the data availability.