API Error Rate and Error Budget, Worked Out
Error rate is failed requests ÷ total requests. A 99.9% SLO on 50 million requests allows 50,000 failures, and a 0.14% day burns that budget 1.4× too fast.
The Calcylator Blog
Practical guides, clear examples, and smarter ways to calculate.
18 articles
Error rate is failed requests ÷ total requests. A 99.9% SLO on 50 million requests allows 50,000 failures, and a 0.14% day burns that budget 1.4× too fast.
Multiply bitrate by concurrent viewers: 1,000 people watching at 5 Mbps need 5 Gbps of delivery and move about 2.25 TB in an hour. Add headroom on top.
Base64 turns every 3 bytes into 4 characters, so encoded length is 4 × ceil(n ÷ 3). One million bytes becomes 1,333,336 characters, about 33% more.
Convert binary to decimal by adding the place values of each 1: 101101 is 32 + 8 + 4 + 1 = 45. Includes the doubling shortcut and an IP address example.
Hit rate is hits ÷ (hits + misses). With a 2 ms cache and 120 ms origin, 92% gives 11.4 ms on average, while 99% gives only 3.2 ms.
A /26 holds 64 addresses and 62 usable hosts. See how 2^(32−n) works, how to find the network and broadcast address, and how to size a subnet.
Explore our free calculators and apply these concepts to your own numbers.