Updated on 2025-07-17 GMT+08:00

Cluster View

A cluster view is generated based on metrics and Prometheus Query Language (PromQL). It displays information about pods, containers, CPUs, memory, networks, and disks, allowing you to better monitor the running of a cluster. The following describes cluster resources from two parts: metric description and metric list. In the resource diagrams, larger values in bytes can be converted to ones in MB, KB, or GB.

Metric Description

The cluster view displays basic resource metrics, network metrics, and disk metrics.

Figure 1 Basic resource diagram
Table 1 Basic resource metrics

Metric

Unit

Description

Pods

Count

Number of pods in different states (Failed, Pending, Running, Succeeded, and Unknown) in a cluster

Containers

Count

Number of containers in different states (Containers Running, Containers Waiting, and Containers Terminated) in a cluster

Used CPU

Core

Total number of CPUs used by all containers in each namespace

Used Memory

Byte

Total amount of memory used by all containers in each namespace

Figure 2 Network diagram
Table 2 Network metrics

Metric

Unit

Description

Network Receive Rate

Byte/s

Total number of bytes received by all containers in each namespace per second

Network Transmit Rate

Byte/s

Total number of bytes sent by all containers in each namespace per second

Average Network Receive Rate

Byte/s

Average number of bytes received by containers in each namespace per second

Average Network Transmit Rate

Byte/s

Average number of bytes sent by containers in each namespace per second

Rate of Received Packets

Count/s

Total number of packets received by all containers in each namespace per second

Rate of Transmitted Packets

Count/s

Total number of packets sent by all containers in each namespace per second

Packet Loss Rate (Receive)

Count/s

Total number of packets not received by all containers in each namespace per second

Packet Loss Rate (Transmit)

Count/s

Total number of packets sent from all containers but not received by the recipient in each namespace per second

Figure 3 Disk diagram
Table 3 Disk metrics

Metric

Unit

Description

Disk I/O Rate (Read + Write)

Count/s

Total number of read and write operations performed on the disk of all containers in each namespace per second

Disk Throughput (Read + Write)

Byte/s

Total number of bytes read from and written to the disk of all containers in each namespace per second

Metric List

The following table lists the metrics used in the cluster view.

Table 4 Cluster metrics

Metric

Type

Description

kube_pod_container_resource_requests

gauge

Container resource requests

kube_pod_container_resource_limits

gauge

Container resource limits

kube_pod_status_phase

gauge

Phase of the pod

kube_pod_container_status_waiting

gauge

Whether the container is in the waiting state

kube_pod_container_status_terminated

gauge

Whether the container is in the terminated state

container_cpu_usage_seconds_total

counter

Cumulative CPU time consumed in seconds

container_memory_rss

gauge

Resident set size (RSS), which is the amount of space of physical memory (RAM) held by a process.

container_network_receive_bytes_total

counter

Cumulative count of bytes received

container_network_transmit_bytes_total

counter

Cumulative count of bytes transmitted

container_network_receive_packets_total

counter

Cumulative count of packets received

container_network_transmit_packets_total

counter

Cumulative count of packets transmitted

container_network_receive_packets_dropped_total

counter

Cumulative count of packets dropped while receiving

container_network_transmit_packets_dropped_total

counter

Cumulative count of packets dropped while transmitting

container_fs_reads_total

counter

Cumulative count of reads completed

container_fs_reads_bytes_total

counter

Cumulative count of bytes read