Spot Instance Savings Calculator
The saving from running a workload on spot (preemptible) instances instead of on-demand — after the work lost to interruptions is re-run — and the interruption rate at which the saving disappears.
Spot capacity sells at a 60–90% discount but can be reclaimed at short notice.
How the spot instance savings calculator works
Spot capacity sells at a 60–90% discount but can be reclaimed at short notice. If a job is interrupted it must redo the work since its last checkpoint, so the effective spot price is the discounted price divided by the useful share of the hours. The saving is what on-demand would cost less that. Checkpointing often, or using many small instances, keeps the lost share small.
Formula: effective spot cost = hours × spot price ÷ (1 − lost share); lost share = interruptions × hours lost per interruption ÷ hours
Worked examples
| Inputs | Saving per month | Note |
|---|---|---|
| 2,000 h at 0.40, 70% off, 2 interruptions per 100 h losing 15 min | 558.79 | saves about 559 a month |
| Frequent interruptions with hourly checkpoints | 500 | saves 500 |
| Interruptions that erase the saving | -1,000 | spot costs more |
FAQFrequently asked questions
What workloads suit spot?
Anything that can stop and resume: batch processing, CI builds, rendering, training with checkpoints, stateless web tier behind a load balancer. Databases and anything with long uncheckpointed state do not.
How often are spot instances interrupted?
It varies by instance type, region and time: many pools see under 5% of instances interrupted per month, popular GPU types far more. Providers publish interruption-frequency bands; diversify across types to reduce it.
Why does checkpointing matter so much?
Because the cost of an interruption is the work since the last checkpoint. Checkpointing every 5 minutes instead of every hour cuts the lost work twelvefold and makes even frequent interruptions cheap.
Is there a middle option?
Reserved or committed-use discounts (30–60%) for steady load, spot for the flexible part, on-demand for the peaks. The reserved-instance break-even calculator covers the first.
Where these figures come from
- IEC 80000-13 — Information science and technology (quantities and units) — the decimal (kB, MB) versus binary (KiB, MiB) prefixes used throughout
- RFC 4632 — Classless Inter-domain Routing (CIDR) — the address-plan arithmetic behind the subnet calculator
- RFC 1918 — Address Allocation for Private Internets — the private ranges the subnet calculator recognises
- NIST SP 800-63B — Digital Identity Guidelines, Authentication — length over composition rules; the basis of the password guidance here
- NIST SP 800-57 Part 1 — Recommendation for Key Management — key-strength comparisons used by the key-space calculator
- Australian Cyber Security Centre — Essential Eight — national guidance on passphrases and system hardening
Last checked: September 2026. Units follow the SI decimal convention (IEC 80000-13 defines the binary alternatives); network and security figures cite the defining standard.