Score a data centre fabric out of 100 on utilisation, oversubscription, availability and drop rate.
Each dimension is normalised to 0–100 with a penalty slope that reflects operational experience — peak utilisation above 60% is where microbursts start to cause drops, and every point of oversubscription beyond non-blocking costs headroom. The weights put availability and utilisation ahead of the counters that merely hint at trouble. It turns a wall of interface counters into one trend line that a capacity review can act on, and it names the single highest-value remediation.
Data Center Network Health Score
score = 0.25 × utilisation + 0.2 × oversubscription + 0.25 × availability + 0.2 × drop rate + 0.1 × port error sub-scores.
score = 0.25 × utilisation + 0.2 × oversubscription + 0.25 × availability + 0.2 × drop rate + 0.1 × port error sub-scores. Each dimension is normalised to 0–100 with a penalty slope that reflects operational experience — peak utilisation above 60% is where microbursts start to cause drops, and every point of oversubscription beyond non-blocking costs headroom. The weights put availability and utilisation ahead of the counters that merely hint at trouble.
It turns a wall of interface counters into one trend line that a capacity review can act on, and it names the single highest-value remediation.
This calculator takes 5 inputs: Peak fabric utilisation, Oversubscription ratio (x:1), Fabric availability over the period, Packet drops from congestion, Ports reporting CRC or link errors. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.