tencent cloud

Feedback

Performance Trends

Last updated: 2022-06-24 16:34:16

    Feature Description

    The performance trends feature monitors the key performance metrics of database instances, Redis nodes, and proxy nodes in real time, including CPU, memory, key information, network usage, network utilization, requests, and responses. It collects monitoring data at the second granularity, dynamically displays the change trends of metrics as well as their maximum, minimum, and average values in tables graphically. You can compare the metrics of multiple nodes and different time periods and zoom in on and drag the monitoring view as needed.
    The powerful statistical analysis capabilities, diversified display methods, and extremely high real-timeness of the performance trends feature can meet your needs in various routine Ops and troubleshooting scenarios of database instances. They help you quickly get a holistic picture of the database performance and prevent risks.

    Monitoring Metrics

    Supported monitoring metrics are displayed in three dimensions: instance, Redis node, and proxy node.

    Instance

    Metric CategoryMetricParameterUnitDescription
    CPU CPU utilization cpu_util%Average CPU utilization
    Max CPU utilization of nodecpu_max_util%The maximum among all node (shard or replica) CPU utilizations in an instance
    Memory information Used memorymem_usedMBActually used memory capacity, including the capacity for data and cache
    Memory utilizationmem_util%The ratio of the actually used memory to the requested total memory
    Max memory utilization of nodemem_max_util%The maximum among all node (shard or replica) memory utilizations in an instance
    Key information Total keyskeys-The total number of keys (level-1 keys) stored in the instance
    Expired keysexpired-The number of keys expired in a time window, which is equal to the value of `expired_keys` output by the `info` command
    Evicted keysevicted-The number of keys evicted in a time window, which is equal to the value of `evicted_keys` output by the `info` command
    Network usage Connectionsconnections-The number of TCP connections to an instance
    Inbound trafficin_flowMB/sPrivate inbound traffic
    Outbound trafficout_flowMB/sPrivate outbound traffic
    Network utilization Connection utilizationconnections_util%The ratio of the number of TCP connections to the maximum number of connections
    Inbound traffic utilizationin_bandwidth_util%The ratio of the actually used private inbound traffic to the maximum traffic
    Outbound traffic utilizationout_bandwidth_util%The ratio of the actually used private outbound traffic to the maximum traffic
    Request Total requestscommandscounts/secQPS, that is, the number of command executions per second
    Read requests cmd_readcounts/secThe number of read command executions per second
    Write requestscmd_writecounts/secThe number of write command executions per second
    Other requests cmd_othercounts/secThe number of command (excluding write and read commands) executions per second
    Big value requests cmd_big_valuecounts/secThe number of executions of requests larger than 32 KB per second
    Key requests cmd_key_countcounts/secThe number of keys requested per second
    MGET executions cmd_cmgetcounts/secThe number of requests made through MGET per second
    Response Slow queriescmd_slow-The number of command executions with a latency greater than the configured `slowlog-log-slower-than` value
    Read request hits cmd_hits-The number of keys successfully requested by read commands, which is equal to the value of the `keyspace_hits` metric output by the `info` command
    Read request misses cmd_miss-The number of keys unsuccessfully requested by read commands, which is equal to the value of the `keyspace_misses` metric output by the `info` command
    Read request hit ratecmd_hits_ratio%Key hits/(key hits + key misses). This metric reflects cache misses. When the access request quantity is 0, the value of this metric will be null
    Execution error Execution errors cmd_err-The number of command execution errors. For example, the command does not exist, or parameters are incorrect.
    Latency Average execution latencylatency_avgmsThe average execution latency between the proxy and the Redis server
    Max execution latencylatency_maxmsThe maximum execution latency between the proxy and the Redis server
    P99 execution latencylatency_p99msThe 99th percentile execution latency between the proxy and the Redis server
    Average read latencylatency_readmsThe average execution latency of read commands between the proxy and the Redis server. For more information on read command types, see Command types.
    Average write latencylatency_writemsThe average execution latency of write commands between the proxy and the Redis server. For more information on write command types, see Command types.
    Average latency of other commandslatency_othermsThe average execution latency of commands (excluding write and read commands) between the proxy and the Redis server

    Redis node

    Metric CategoryMetricParameterUnitDescription
    CPU CPU utilizationcpu_util%Average CPU utilization
    Network usage Connectionsconnections-The number of connections between the proxy and a node
    Connection utilizationconnections_util%The connection utilization of a node
    Memory information Used memorymem_usedMBActually used memory capacity, including the capacity for data and cache
    Memory utilizationmem_util%The ratio of the actually used memory to the requested total memory
    Key information Total keyskeys-The total number of keys (level-1 keys) stored in the instance
    Expired keysexpired-The number of keys expired in a time window, which is equal to the value of `expired_keys` output by the `info` command
    Evicted keysevicted-The number of keys evicted in a time window, which is equal to the value of `evicted_keys` output by the `info` command
    Replication delayReplication delayrepl_delayByteThe command delay between the replica node and the master node
    Request Total requestscommandscounts/secQPS, that is, the number of command executions per second
    Read requestscmd_readcounts/secThe number of read command executions per second. For more information on read command types, see Command types.
    Write requestscmd_writecounts/secThe number of write command executions per second. For more information on write command types, see Command types.
    Other requestscmd_othercounts/secThe number of command (excluding write and read commands) executions per second
    Response Slow queriescmd_slow-The number of command executions with a latency greater than the configured `slowlog-log-slower-than` value
    Read request hitscmd_hits-The number of keys successfully requested by read commands, which is equal to the value of the `keyspace_hits` metric output by the `info` command
    Read request missescmd_miss-The number of keys unsuccessfully requested by read commands, which is equal to the value of the `keyspace_misses` metric output by the `info` command
    Read request hit ratecmd_hits_ratio%Key hits/(key hits + key misses). This metric reflects cache misses.

    Proxy node

    Metric CategoryMetricParameterUnitDescription
    CPU CPU utilizationcpu_util%Proxy CPU utilization
    Request Total requestsproxy_commandscounts/secThe number of proxy command executions per second
    Key requestscmd_key_countkeys/secondThe number of keys accessed by a command per second
    MGET requestscmd_mgetcounts/secThe number of MGET command executions per second
    Execution errorscmd_errcounts/secThe number of proxy command execution errors per second. For example, the command does not exist, or parameters are incorrect.
    Big value requestscmd_big_valuecounts/secThe number of executions of requests larger than 32 KB per second
    Traffic Inbound trafficin_flowMB/sPrivate inbound traffic
    Outbound trafficout_flowMB/sPrivate outbound traffic
    Network usage Connectionsconnections-The number of TCP connections to an instance
    Connections per secclient_connections_received_per_second-The number of TCP connections established per second
    Disconnections per secondclient_connections_closed_per_second-The number of TCP connections closed per second
    Abnormal disconnections per secclient_connections_aborted_per_second-The number of TCP connections aborted per second
    Network utilization Connection utilizationconnections_util%The ratio of the number of TCP connections to the maximum number of connections
    Inbound traffic utilizationin_bandwidth_util%The ratio of the actually used private inbound traffic to the maximum traffic
    Inbound traffic limit countin_flow_limit-The number of times inbound traffic triggers a traffic limit
    Outbound traffic utilizationout_bandwidth_util%The ratio of the actually used private outbound traffic to the maximum traffic
    Outbound traffic limit countout_flow_limit-The number of times outbound traffic triggers a traffic limit
    Latency Average execution latencylatency_avgmsThe average execution latency between the proxy and the Redis server
    Max execution latencylatency_maxmsThe maximum execution latency between the proxy and the Redis server
    P99 execution latencylatency_p99msThe 99th percentile execution latency between the proxy and the Redis server
    Average read latencylatency_readmsThe average execution latency of read commands between the proxy and the Redis server. For more information on read command types, see Command types.
    Average write latencylatency_writemsThe average execution latency of write commands between the proxy and the Redis server. For more information on write command types, see Command types.
    Average latency of other commandslatency_othermsThe average execution latency of commands (excluding write and read commands) between the proxy and the Redis server

    Viewing Monitoring Data

    Step 1. Select monitoring metrics

    1. Log in to the TencentDB for Redis console.
    2. On the left sidebar, select Performance Optimization.
    3. At the top of the Performance Optimization page of DBbrain, select the target instance in the Instance ID drop-down list.
    4. Click the Performance Trends tab, select the target performance metrics in the metric category drop-down list, and click Save.
      To apply the selected performance metrics to all the TencentDB for Redis instances under your account, click Save and Apply to All Instances as shown below:

    Step 2. Set the collection granularity

    In the top-right corner of the Performance Trends tab, set the collection granularity of the monitoring data in the drop-down list on the right of Auto-Refresh to 5s, 15s, or 30s as shown below:

    Viewing monitoring metrics in different dimensions

    Below the metric categories on the Performance Trends tab, you can view the monitoring metric data by instance, Redis node, and proxy node as needed.

    img

    Comparing the performance metrics of multiple nodes

    1. On the Performance Trends tab, click Multi-Node Performance Comparison.
    2. In the Multi-Node Performance Comparison panel, click Create Multi-Node Performance Comparison Task.
    3. In the Create Multi-Node Performance Comparison Task window, click in the selection box next to Monitoring Time to select the monitoring time period, select the target monitoring metric in the Monitoring Metric drop-down list, and click OK.
    4. Wait for the Status to become Successful in the task list in the Multi-Node Performance Comparison panel.
    5. Click View in the Operation column to view the comparison data of all Redis nodes. The connections metric is used as an example as shown below:

    Switching between real-time and historical views

    On the Performance Trends tab, the real-time monitoring data is displayed by default.

    • In routine Ops monitoring, database instance metrics can be monitored in real time.
    • When you need to locate exceptions, you can click Historical to analyze the monitoring data in a past time period.
      • The monitoring data in the last 1 hour, 3 hours, and 7 days can be viewed.
      • Click to view the monitoring data in any time period in the last 30 days.

    Comparing performance metrics in different time periods

    1. On the Performance Trends tab, click Historical and then click Add Time Comparison.
    2. In the time selection box, select two time periods for comparison.
    3. Select the target monitoring metrics and hover over the change trend in the monitoring view to compare the monitoring data in the two time periods.

    Displaying monitoring metric data in chart

    • Click next to Show Statistics as shown below to display the max, min, and average values of each monitoring metric in a table.
    • Click in the top-right corner of any monitoring view to display the max, min, and average values of the monitoring metric in a table.
      The Network Usage metric is used as an example as shown below:

    Viewing monitoring data through chart interaction

    The Chart Interaction feature is suitable for analyzing the data of a monitoring view and its associated monitoring views.

    1. In the top-right corner of the Performance Trends tab, click next to Chart Interaction.
    2. In any of the monitoring views to be analyzed, select a time point and click it, and the data at the same time point will be fixed for display in other monitoring views.
    3. You can click Deselect the Time Point in the top-right corner of the monitoring view to cancel the fixed display.
      img

    Customizing monitoring metric for comparative analysis

    Click in the top-right corner of any monitoring view to add monitoring metrics of other types for comparative display and analysis.

    Switching between one-column and two-column mode of monitoring view

    Click on the right of Chart Interaction in the top-right corner to switch between the one-column and two-column modes. The former is as shown below:

    Dragging monitoring view

    The monitoring views can be freely dragged to flexible adjust their order for efficient display and analysis.

    Zooming in on monitoring view

    Drag the icon in the bottom-right corner of any monitoring view to zoom in on the image for clearer display of the metric trends.

    Contact Us

    Contact our sales team or business advisors to help your business.

    Technical Support

    Open a ticket if you're looking for further assistance. Our Ticket is 7x24 avaliable.

    7x24 Phone Support