Skip to main content
Version: User Guides (Cloud)

View Cluster Metric Charts

Zilliz Cloud offers a dashboard for observing cluster-specific metrics. To access this feature, navigate to the Metrics tab within one of your clusters.

📘Notes

This feature is exclusively available to clusters in the Standard and Enterprise plans. For more information, see Select the Right Cluster Plan.

Access cluster metric charts​

In the Zilliz Cloud console, locate the target cluster and select the Metrics tab.

Zilliz Cloud's metric charts provide performance data on resource usage, queries per second (QPS), request results, and data operations, offering granular analysis within a specific time range.

📘Notes

Clicking View Alerts Settings on the right will redirect you to the Alert Settings page, offering a shortcut to manage your alerts.

view_metric_charts_resources

For details on each metric chart, refer to View metric charts.

Modify curve window size​

The Metrics tab allows for two types of window sizes.

  • Relative Range: Choose from a set of pre-defined time periods relative to your current time. Using relative time ranges allows you to check metrics in a periodical and convenient way, without needing to enter the specific start and end time. Your choices include:

    • Last 10 minutes

    • Last hour

    • Last day

    • Last week

    • Last month

  • Absolute Range: Enter the exact start time and end time. Using absolute range allows you to control the metrics you see in a more fine-tuned way.

    • The time difference between the start and the end time should be greater than 10 minutes.

filter_metrics_by_time_period

View metric charts​

Zilliz Cloud offers metric charts for monitoring cluster performance from various aspects. For a quick overview of available metrics, refer to Metrics & Alerts Reference.

Resources​

To view metric charts for resource usage, select the Metrics tab and refer to the Resources area. These charts provide a snapshot of the cluster's resource usage, including computation, capacity, and storage.

  • CU Computation: Measures computational power usage relative to the total.

    • 70%-80%: Prepare for scaling out.

    • 90% or more: Scale out immediately to avoid service disruptions.

  • CU Capacity: Indicates used capacity as a percentage of total, measured per second over time.

    • 70%-80%: Prepare for scaling out.

    • 90%-99%: Scale out immediately to avoid service disruptions.

    • 100%: When the used CU capacity hits 100%, Zilliz Cloud disables data writing and triggers SDK errors. To restore normal functionality, scale out your cluster immediately.

  • Storage Use: Shows total persistent storage consumed, calculated in GB/s over a selected period.

view_metric_charts_resources

Performance​

To view metric charts for performance, select the Metrics tab and refer to the Performance area. These charts provide a snapshot of cluster performance, including QPS, VPS, latency, and request .

  • QPS/VPS (Read)

    • QPS: The number of read requests (search and query) per second.

    • VPS: The number of read requests (search) on vectors per second. VPS is not available for query requests as query operations do not involve vectors.

  • QPS/VPS (Write)

    • QPS: The number of write requests (insert, bulk insert, upset, and delete) per second.

    • VPS: The number of write requests (insert, bulk insert,upset, and delete) on vectors per second.

  • Latency (Read): The time elapsed between a client sending a read request (search and query request) to a server and the client receiving a response. It includes an average latency and a P99 latency.

  • Latency (Write): The time elapsed between a client sending a write request (insert and upsert request) to a server and the client receiving a response. It includes an average latency and a P99 latency.

  • Request Failure Rate (Read): The percentage of timeout read requests in all requests per second.

  • Request Failure Rate (Write): The percentage of timeout write requests in all requests per second.

view_metric_charts_performance

Data​

To view metric charts for business data, select the Metrics tab and refer to the Data area. These charts provide a snapshot of the cluster's entity data by indicating the number of collections, entities, and loaded entities in the cluster.

  • Collection Count: The count of collections created in the cluster.

  • Entity Count: The count of entities inserted into the cluster.

  • Loaded Entities: The count of entities loaded in the cluster.

view_metric_charts_entity