SMB overview dashboard

Learn about the SMB overview dashboard.

The SMB overview dashboard provides a real-time visualization of SMB (Server Message Block) protocol activity and performance across your IBM Storage Ceph cluster. SMB is commonly used for file sharing in networked environments, and this dashboard enables you to monitor key performance indicators such as throughput, latency, session activity, and I/O operations.This dashboard is intended for system administrators, performance engineers, and support personnel responsible for maintaining high availability and performance of SMB services.

The dashboard also provides per-share performance details depending on the selected share or filters. These values appear across multiple graphs rather than in a single panel. Per-share indicators include:
Throughput
Data transfer rate per share (kB/s, MB/s, or GB/s).
IOPS
Number of input/output operations per second per share.
Latency
Average response time per share, measured in microseconds (µs).

Ensure that the monitoring stack is installed. For more information, see Deploying the monitoring stack.

Go to the Ceph dashboard > File > SMB > Overview to view the dashboard. You can also access the dashboard from Ceph dashboard > Administration > Services > Grafana > SMB Overview.

To use the dashboard effectively,

  • Use the time range selector in the upper-right to adjust the visibility window for metrics.
  • Use filters such as Ceph cluster, SMB Cluster, Share , and Client to go into specific nodes or clusters.
  • Look for latency and I/O spikes, which can indicate performance degradation or high usage periods.
  • Refer to this dashboard during troubleshooting, capacity planning, and performance tuning.
Note: The Client option is available only as a filter. The dashboard does not display client counts as a separate metric.
The following lists the dashboard pages and descriptions:
SMB metrics status
The operational status of the Prometheus exporter collecting SMB metrics. A green Up status means that the exporter is running and metrics are actively used.
SMB active sessions
The number of currently active SMB sessions in the cluster.
SMB shares activity
The current activity across all SMB shares within the cluster.
SMB ingress throughput
The rate at which data is received by the SMB server, which is measured in bytes (B). Helps monitor incoming data load from clients.
SMB latency
The average latency for SMB operations, which are measured in microseconds (µs). Useful for identifying performance issues; lower latency values indicate better responsiveness.
SMB egress throughput
The rate at which data is sent out from the SMB server, which is measured in bytes (B).
SMB I/O
The number of SMB input/output operations per second across all nodes. Shows a quick view of overall I/O activity.
SMB I/O per Operation (Graph)
Displays the number of operations per second broken down by SMB operation type (for example: create, read, write, flush, close). This helps identify which SMB operations are generating the most load and how workloads vary over time.
SMB throughput (Graph)
A time series graph plotting ingress and egress throughput across the monitored period. Helps identify usage patterns, peaks, and anomalies over time.
SMB latencies (Graph)
A time series visualization of SMB operation latencies (for example, create, close, flush) for each node.