Edge Performance Score Calculator

Score an edge delivery setup on cache efficiency, tail latency, reliability and protocol adoption, then rank the weakest.

Inputs

Byte cache hit ratio%

Observed p95 response timems

p95 latency targetms

5xx error rate at the edge%

Measured availability%

Handshakes resumed or 0-RTT%

Compressible responses actually compressed%

Traffic served over HTTP/3%

Overall Edge Score

77.7/ 100

Grade

B — Good

Cache Efficiency Pillar

92.6/ 100

Tail Latency Pillar

55.6/ 100

Reliability Pillar

93.8/ 100

Protocol Adoption Pillar

65.0/ 100

Fix This First

Tail latency — move processing to the edge and shrink the largest responses

Step by step

Values used
Byte cache hit ratio = 88 %; Observed p95 response time = 180 ms; p95 latency target = 100 ms; 5xx error rate at the edge = 0.4000 %; Measured availability = 99.95 %; Handshakes resumed or 0-RTT = 65 %; Compressible responses actually compressed = 80 %; Traffic served over HTTP/3 = 45 %
Edge Performance Score
score = cache × 0.30 + tail latency × 0.30 + reliability × 0.25 + protocol adoption × 0.15, where cache is scored against a 95% target and latency as target ÷ observed p95.
Overall Edge Score
= 77.7 / 100
Grade
= B — Good
Cache Efficiency Pillar
= 92.6 / 100
Tail Latency Pillar
= 55.6 / 100
Reliability Pillar
= 93.8 / 100
Protocol Adoption Pillar
= 65.0 / 100

How it works

Each pillar is normalised to 100 from a figure you can pull from CDN analytics, with cache and tail latency carrying the most weight because they dominate what users experience. Latency scores as a ratio rather than a subtraction, so a p95 at twice the target scores 50 regardless of the absolute numbers. A single score makes edge configuration comparable across properties and over time, and naming the weakest pillar turns it into one concrete action instead of a dashboard nobody acts on.

Formula

Edge Performance Score

score = cache × 0.30 + tail latency × 0.30 + reliability × 0.25 + protocol adoption × 0.15, where cache is scored against a 95% target and latency as target ÷ observed p95.

cache pillar: Byte hit ratio scored against a 95% practical ceiling
latency pillar: Latency target divided by the observed p95, capped at 100
reliability pillar: Error rate penalised at 20 points per percent, blended with availability

Frequently Asked Questions

How is Edge Performance Score calculated?

score = cache × 0.30 + tail latency × 0.30 + reliability × 0.25 + protocol adoption × 0.15, where cache is scored against a 95% target and latency as target ÷ observed p95. Each pillar is normalised to 100 from a figure you can pull from CDN analytics, with cache and tail latency carrying the most weight because they dominate what users experience. Latency scores as a ratio rather than a subtraction, so a p95 at twice the target scores 50 regardless of the absolute numbers.

Why does Edge Performance Score matter?

A single score makes edge configuration comparable across properties and over time, and naming the weakest pillar turns it into one concrete action instead of a dashboard nobody acts on.

What values do I need to enter?

This calculator takes 8 inputs: Byte cache hit ratio, Observed p95 response time, p95 latency target, 5xx error rate at the edge, Measured availability, Handshakes resumed or 0-RTT, Compressible responses actually compressed, Traffic served over HTTP/3. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.

You might also need

CDN Cache Hit Ratio Calculator API Response Latency Calculator Edge Location Planner