The cache analysis feature can analyze backup files of ApsaraDB for Redis and identify big keys in your instances. This feature allows you to view the information about an instance, such as the memory usage, distribution, and expiration time of keys. You can optimize your instances based on the analysis results. This feature helps you resolve issues such as insufficient memory and performance degradation that are caused by skewed distribution of keys.

Prerequisites

An ApsaraDB for Redis Community Edition or Enhanced Edition instance is created. For more information, see Performance-enhanced instances, Persistent memory-optimized instances, and Storage-optimized instances.

Procedure

  1. Log on to the ApsaraDB for Redis console.
  2. In the top navigation bar of the page, select the region where the instance is deployed.
  3. On the Instances page, click the ID of the instance.
  4. In the left-side navigation pane, choose CloudDBA > Cache analysis.
    By default, the analysis results of the last day are displayed on the Cache analysis tab. You can specify another time range based on your requirements.
  5. On the Cache analysis page, click Analyze.
    Cache analysis
  6. In the dialog box that appears, specify the node and analysis method.
    Specify the node and analysis method
    Parameter Description
    Select a node for analysis The ID of the node on which you want to perform cache analysis.
    Note You can select an instance or a node for analysis. If you select an instance that contains more than eight nodes, the system analyzes only the top eight nodes that have the highest memory usage.
    Analysis Method The method that you want to use to analyze the cache. Valid values: Use Previous Backup File and Create a new backup and use the latest backup for analysis.
  7. Click OK.
    The system performs cache analysis and displays the analysis status. You can click Refresh to update the analysis status.
  8. Find the completed analysis task and click Details in the Actions column to view the results.
    Note If you select a cluster instance or a read/write splitting instance from the Select an analysis node drop-down list, you can view the information about Related nodes and select a node in the Details section of the Details page.
    • Basic information: displays basic information about an instance such as the basic attributes and cache analysis method. Basic information
    • Related node: displays the memory usage and statistics about keys on each node of the instance. Result 1
    • Details: displays the details about an instance or a node, such as the memory usage, distribution of keys, memory usage and distribution of elements in keys, distribution of key expiration time, and ranking of big keys. Details of cache analysis

FAQ

Q: Why do I fail to perform cache analysis?
Figure 1. Failed cache analysis
Failed cache analysis

A: Bloom filter-related API operations, such as BF.ADD, on existing ApsaraDB for Redis 4.0 instances are only for internal use. These API operations are no longer supported by new ApsaraDB for Redis 4.0 and later instances. Therefore, if you call Bloom filter-related API operations, cache analysis fails and unknown errors may occur. We recommend that you change your instances to Performance-enhanced instances of ApsaraDB for Redis Enhanced Edition that support the optimized Bloom filter.