Cloud Log Service (CLS) is a SaaS service with integrated observability that supports Log and Metric data collection, ETL, search and analysis, visualization, and alerting.
The common terms used in CLS APIs are listed in the following table:
| Term | Description |
|---|---|
| Logset | Logset is a classification of log topics and metric topics. A logset can contain multiple log topics and metric topics. The logset itself does not store any log data, only making user management convenient. A logset usually corresponds to a certain project/business in the company. It is advisable to group log topics of multiple applications/services under the same logset. |
| Log Topic | A log topic is the basic unit for log data collection, storage, retrieval, and analysis on the CLS platform. Massive logs are managed by log topic, including collection rule configuration, retention time configuration, log retrieval analysis, and log download/consumption/delivery. |
| Metric Topic | A metric topic is the basic unit for metric data collection, storage, retrieval, and analysis on the CLS platform. Collected metric data is managed by metric topic, including retention time configuration and retrieval analysis. |
| Machine Group | A machine group is a list of machines needing log collection. Log Service uses machine groups to manage all machines that need to collect logs using LogListener. |
| Topic Partition | A topic partition is the smallest read/write unit in CLS. A log topic can contain multiple partitions, and the number of partitions determines the maximum data read/write capacity of a single log topic. |
| Data Processing | Data processing is the process of filtering, cleaning, masking, enriching, and distributing log data to target log topics. It can be understood as log ETL (Extract-Transform-Load). |
| Scheduled SQL Analysis | Scheduled SQL analysis is the process of periodically querying log data (support retrieval and SQL) based on a specified time window and saving query results to target log topic. |
| Log Group | A log group (LogGroup) is a collection of multiple logs. For log upload, to improve data reading and writing efficiency, multiple logs are packaged into a log group and uploaded to CLS in units of log groups. Logs in a log group have the same basic information (TIMESTAMP, FILENAME, SOURCE, HOSTNAME, and metadata). |
| Metric | A metric is a measurement used to evaluate the performance and running status of systems and applications, such as CPU utilization, memory usage, access throughput, response latency, and response success rate. Metrics are usually generated periodically, with a value at each moment, forming a sequence over time. This sequence is generally called a time series. It is used to collect and record data in systems and applications. |
For usage limits, please refer to Document Center > Log Service > Product Introduction > Specifications and Limits.
You can use the API Explorer tool to call APIs online.
This document uses creating a log topic - uploading logs - retrieving logs as an example. The steps to call the API via the API Explorer Tool are as follows:
フィードバック