In short
- Definition: The ability of a system to absorb disruptions, continue operating in a degraded mode, and recover quickly.
- Goal: Keep essential services running under stress and return to normal fast.
- Example S.M.A.R.T. requirement: Core functions available during a dependency outage, and full recovery within 15 minutes after it ends.
1. Definition
As a non-functional requirement, resilience describes the ability of a system to absorb disruptions, keep its essential functions running in a degraded mode, and return to normal quickly [1], [2].
A resilient system bends instead of breaking, and knows how to straighten up again.Key takeaway
Disruptions include component failures, overloads, dependency outages, bad deployments and attacks. Resilient systems use timeouts, retries with backoff, circuit breakers, bulkheads that isolate failures, graceful degradation and automated recovery. The AWS Well-Architected reliability pillar, for example, lists automatic recovery from failure and testing recovery procedures among its design principles [3].
Resilience goes beyond fault tolerance: it accepts that some failures will affect users and focuses on limiting and shortening their impact.
2. Goal
The goal is to keep essential services running under stress and to return to normal fast.
Without resilience, small problems cascade into full outages and recovery depends on manual heroics.
3. How to Measure It (S.M.A.R.T.)
| Metric | Unit | Typical target | How to measure |
|---|---|---|---|
| Core functions available during a dependency outage | % of core functions | 100% | Failure injection tests |
| Time to full recovery after disruption | Minutes | < 15 | Incident and test records |
| Cascading failures | Count per quarter | 0 | Incident analysis |
| Resilience experiments | Per month | ≥ 1 | Chaos engineering log |
4. Example Requirements
Each requirement below is ready to copy into a specification: specific, measurable, achievable, relevant and time-bound.
- Degraded mode: “When the payment provider is unavailable, customers shall still be able to browse, fill their cart and save orders for later payment.”
- Recovery: “After any dependency outage ends, the service shall return to full function within 15 minutes without manual action.”
- Isolation: “Overload of the reporting module shall not affect response times of the ordering module, as verified by monthly load and chaos tests.” [4]
5. Questions to Ask Stakeholders
- Which functions are essential and must survive disruptions?
- Which degraded modes are acceptable to users and the business?
- Which dependencies are most likely to fail?
- How fast must the system return to normal?
- How will resilience be tested?
- Who decides to switch to a degraded mode?
6. Common Pitfalls
- Retries without limits. They amplify overloads. Use backoff, jitter and limits.
- No timeouts. A slow dependency can hold every thread. Always set timeouts.
- Degraded modes nobody tested. Fallback paths must be exercised to work when needed.
7. Related NFRs
- Fault Tolerance — masking component failures entirely.
- Availability — resilience protects uptime and shortens outages.
- Disaster Recovery — handles severe events beyond normal resilience.
- Failure Management — detects and responds to failures.
- Robustness — withstands bad inputs and abnormal conditions.
8. FAQ
Fault tolerance aims for no visible impact from a failure; resilience limits the impact and recovers quickly, accepting some degradation.
Keeping core functions working with reduced features when parts of the system fail.
A pattern that stops calls to a failing dependency for a while, so the failure does not spread.
9. Acronyms, Initialisms & Terms
Acronyms are abbreviations pronounced as words (like NASA); initialisms are spelled out letter by letter (like FBI). The classification below follows common pronunciation.
Acronyms
| Abbreviation | Stands for | Definition | Ref. |
|---|---|---|---|
| SMART | Specific, Measurable, Achievable, Relevant, Time-bound | Criteria for writing a goal or requirement so that anyone can check whether it has been met. | [5] |
Initialisms
| Abbreviation | Stands for | Definition | Ref. |
|---|---|---|---|
| AWS | Amazon Web Services | Amazon’s cloud computing platform. | [6] |
| FAQ | Frequently asked questions | List of common questions with their answers. | [7] |
| NFR | Non-functional requirement | Requirement that describes how well a system must work (its qualities, such as speed or security) rather than what it does. | [2] |
Terms
Plain-language definitions of technical and project-management terms used in this article.
| Term | Plain-language definition | Ref. |
|---|---|---|
| Chaos engineering | Deliberately injecting failures in a controlled way to check that a system copes. | [8] |
| Circuit breaker | Software pattern that stops calling a failing service for a while so failures do not spread. | [9] |
| Deployment | Installing a new version of software in the environment where people use it. | [10] |
| Incident | Unplanned interruption or degradation of a service. | [11] |
| Log | Time-stamped record of events written by software, used for troubleshooting and audit. | [12] |
| Stakeholder | Person or group with an interest in the system: users, sponsors, operators, regulators. | [13] |
| Uptime | Time during which a system is working and available. | [14] |
10. Sources and Further Reading
- [1] “Resilience (network).” Wikipedia. https://en.wikipedia.org/wiki/Resilience_(network) (accessed Nov. 18, 2026).
- [2] “Non-functional requirement.” Wikipedia. https://en.wikipedia.org/wiki/Non-functional_requirement (accessed Nov. 18, 2026).
- [3] Amazon Web Services. “Reliability pillar, AWS Well-Architected Framework.” AWS Documentation. https://docs.aws.amazon.com/wellarchitected/latest/reliability-pillar/welcome.html (accessed Nov. 18, 2026).
- [4] principlesofchaos.org. “Principles of Chaos Engineering.” Principles of Chaos Engineering. https://principlesofchaos.org/ (accessed Nov. 18, 2026).
- [5] “SMART criteria.” Wikipedia. https://en.wikipedia.org/wiki/SMART_criteria (accessed Nov. 19, 2026).
- [6] “Amazon Web Services.” Wikipedia. https://en.wikipedia.org/wiki/Amazon_Web_Services (accessed Nov. 19, 2026).
- [7] “FAQ.” Wikipedia. https://en.wikipedia.org/wiki/FAQ (accessed Nov. 19, 2026).
- [8] “Chaos engineering.” Wikipedia. https://en.wikipedia.org/wiki/Chaos_engineering (accessed Nov. 19, 2026).
- [9] “Circuit breaker design pattern.” Wikipedia. https://en.wikipedia.org/wiki/Circuit_breaker_design_pattern (accessed Nov. 19, 2026).
- [10] “Software deployment.” Wikipedia. https://en.wikipedia.org/wiki/Software_deployment (accessed Nov. 19, 2026).
- [11] “Incident management.” Wikipedia. https://en.wikipedia.org/wiki/Incident_management (accessed Nov. 19, 2026).
- [12] “Logging (computing).” Wikipedia. https://en.wikipedia.org/wiki/Logging_(computing) (accessed Nov. 19, 2026).
- [13] “Project stakeholder.” Wikipedia. https://en.wikipedia.org/wiki/Project_stakeholder (accessed Nov. 19, 2026).
- [14] “Uptime.” Wikipedia. https://en.wikipedia.org/wiki/Uptime (accessed Nov. 19, 2026).
Written by Philippe Huard · LinkedIn · Last updated: November 18, 2026 · Header photo by @ashuba; inline photo by @divineproductionsfilms, both on Unsplash.




