Skip to content

Metrics

View system performance metrics for each batch including memory and CPU usage.

Overview

Access metrics from the batch detail view by selecting Metrics from the Data Type dropdown.

Memory Usage

Chart

Line/Area chart showing memory usage over the batch time window.

Element Description
X-axis Time (HH🇲🇲ss)
Y-axis Usage (0-100%)
Blue Area Memory usage trend

Threshold Indicators

The colours use the same thresholds the backend uses to flag anomalies for the AI, so a red value is one the analysis also treats as abnormal.

Level Memory CPU Color
Over threshold > 75% > 80% 🔴 Red
Approaching > 68% > 72% 🟡 Yellow
Healthy ≤ 68% ≤ 72% 🟢 Green

"Approaching" starts at 90% of the threshold. Reference lines on the chart show these thresholds.

Average Usage

Circular gauge showing average memory usage for the batch with color-coded status.

CPU Usage

Chart

Same format as memory, showing CPU usage over time.

Average Usage

Circular gauge with percentage and threshold indicators.

Filtering

Host Filter

Filter metrics by host when multiple hosts are present: - All Hosts (default) - Individual host names discovered from metrics

Type Filter

Both Memory and CPU usage are displayed simultaneously. Use the type dropdown to scroll to the selected metric type: - Memory Usage - CPU Usage

Action Description
Back Returns to batch detail view
Type Dropdown Switch between metric types
Host Dropdown Filter by specific host

Data Source

Metrics are fetched from the configured Metric Provider (Prometheus, VictoriaMetrics, Datadog, etc.) and stored within batches.

Field Description
timestamp When metric was recorded
host Host/instance identifier
value Usage percentage (0-100)
type memory_usage or cpu_usage