This topic describes performance metrics related to Linux system load on a host. You can determine whether the current host is overloaded based on the trend charts of the metrics: load1, load5, and load15. If the value of load5 or load15 is always greater than the total number of CPU cores on the host, the host is overloaded. You can run the grep 'model name' /proc/cpuinfo | wc -l command to query the number of CPU cores of the current host.
load1
Definition
The average system load in the last minute.
Description
| Metric | Metric name | Unit |
|---|---|---|
| load1 | load1 | N/A |
Calculation expression
avg(node_load1{@LABELS}) by (@GBLABELS)
load5
Definition
The average system load in the last five minutes.
Description
| Metric | Metric name | Unit |
|---|---|---|
| load5 | load5 | N/A |
Calculation expression
avg(node_load5{@LABELS}) by (@GBLABELS)
load15
Definition
The average system load in the last 15 minutes.
Description
| Metric | Metric name | Unit |
|---|---|---|
| load15 | load15 | N/A |
Calculation expression
avg(node_load15{@LABELS}) by (@GBLABELS)
load1_per_cpu
Definition
The ratio between the value of the load1 metric and the total number of CPU cores.
Description
| Metric | Metric name | Unit |
|---|---|---|
| load1_per_cpu | load1_per_cpu | N/A |
Calculation expression
sum(node_load1{@LABELS}) by (@GBLABELS) / sum(cpu_count{@LABELS}) by (@GBLABELS)
