What is dashboards per thread per minute?

The per-thread refresh rate from your analytics server. Profile a representative dashboard set and get a baseline; do not use vendor maximums.

Why include both uptime and SLA share?

Uptime is binary (server up or not). SLA share is graded (refresh completed in time or not). Both are losses; the calculator separates them.

How do I model multi-thread servers?

Multiply dashboards-per-thread by thread count. Most analytics servers run 4 to 16 worker threads in production.

Should I scope to dashboards or to widgets?

Scope to dashboards if dashboards are roughly equivalent in size. If they vary widely, scope to widgets and use widgets-per-thread instead.

IIoT, SCADA & Edge Connectivity calculator

Dashboard Refresh Latency Calculator

Estimate concurrent dashboard refresh capacity. Enter dashboards refreshed per server thread per minute, planned refresh minutes in the period, server uptime, and the share of refreshes inside the latency SLA. The calculator returns gross capacity, dashboards delivered inside SLA, and the loss buckets.

What this calculator does

Estimate concurrent dashboard refresh capacity from dashboards refreshed per server thread per minute, planned refresh minutes in the period, server uptime, and the share of refreshes that complete inside the latency SLA.
Use it when an OT data ops or analytics platform lead is sizing how many dashboards an analytics server (PI Vision, AVEVA Insight, Grafana, Power BI gateway) can keep refreshed cleanly.
It returns the dashboards an analytics server can refresh cleanly inside the latency SLA in the period.

Formula used

Gross dashboard refresh capacity = dashboards per minute × planned minutes
Dashboards delivered inside SLA = gross capacity × server uptime × SLA refresh rate

Inputs explained

Dashboards refreshed per thread per minute: Use the per-thread refresh rate from analytics server profiling (typical 4 to 20 per thread per minute on PI Vision and Grafana, lower on heavy multi-tag dashboards).
Planned refresh minutes in the period: Use planned minutes the analytics server is operating (480 minutes per shift if single-shift, 1440 if continuous).
Analytics server uptime: Use the share of the period the analytics server was actually serving (typical 99+ percent).
Refreshes inside SLA: Use the share of refreshes completing inside the latency SLA (typical 95 to 99 percent for healthy dashboards; lower if backend historian queries are slow).

How to use the result

Use it before scaling the analytics server, when adding a new tile-heavy dashboard to a busy server, or when users complain about slow dashboards in the operations meeting.
Per-thread refresh rate depends heavily on tag count per dashboard, query type, and historian cache. Profile against your actual dashboards before sizing capital.

Common questions

What is dashboards per thread per minute? The per-thread refresh rate from your analytics server. Profile a representative dashboard set and get a baseline; do not use vendor maximums.
Why include both uptime and SLA share? Uptime is binary (server up or not). SLA share is graded (refresh completed in time or not). Both are losses; the calculator separates them.
How do I model multi-thread servers? Multiply dashboards-per-thread by thread count. Most analytics servers run 4 to 16 worker threads in production.
Should I scope to dashboards or to widgets? Scope to dashboards if dashboards are roughly equivalent in size. If they vary widely, scope to widgets and use widgets-per-thread instead.

Last reviewed 2026-05-12.