blast radius
How much of a system is affected when something fails or has to be running for a check to happen.
In more detail
In testing, a test with a large blast radius needs many parts alive and can fail for many unrelated reasons. In operations, the term describes how many users or services a bad change can hurt. Both senses push you towards small, contained checks and changes.
Examples
In AWS: How much harm one failure, mistake or stolen credential can cause before something stops it.A remediation job that refuses to change more than three resources per runblast radius: 25 percent of users, for one waveStarting a canary at 10% of traffic limits a bad release to a tenth of users for a few minutes.The course frames least privilege as a blast radius question: assume the credential leaks, then ask what it can do in the next few minutes.Related terms
- canary release — Sending a small share of real traffic to a new version first, and increasing it only while its metrics stay healthy.
- feature flag — A switch in the code, controlled from outside, that turns a feature on or off for chosen users without a new deployment.
- least privilege — Giving every program and person only the access they need, and nothing more.
- availability zone — One of several isolated groups of data centres inside a cloud region, with separate power and networking so a failure in one does not take down the others.
- guardrail — A control that blocks a dangerous action whatever else is allowed, such as an explicit deny or a block on public access.
- AWS account — The container that holds your AWS resources, identities and bill, identified by a 12-digit account number.
- automated remediation — Fixing a detected misconfiguration automatically, such as removing an open SSH rule as soon as it appears.
- percentage rollout — Releasing a feature to a chosen percentage of users, then raising the percentage as confidence grows.
- rolling update — Replacing instances of an application with the new version a few at a time while the rest keep serving.
- progressive delivery — Releasing a new version to a growing share of users step by step, checking health at each step before going further.
- failure domain — A group of components that can fail together because they share something, such as a rack, power supply or zone.
- fault isolation — Designing so a failure in one part cannot take down the others.
- noisy neighbor — One tenant or client that uses so much of a shared resource that others on the same system suffer.
- test pyramid — A guide for balancing a test suite: many fast unit tests, fewer integration tests and a small number of end-to-end tests.
- end-to-end test — A test that drives the whole system from the outside, often through a browser or public API, with every part running.
Taught in
- Threat-Model An LLM App — AI Engineering: LLM Apps, RAG & Agents
- Canary Releases — Continuous Integration & Continuous Delivery Pipelines
- Policies and Least Privilege — Cloud Infrastructure & Hyperscaler Architecture
- AWS Organizations, Account Hierarchies & Azure Management Groups — Cloud Security & DevSecOps Engineering
- Workload Identity Federation: Eliminating Static Keys — Cloud Security & DevSecOps Engineering
- Cloud Security Posture Management & Remediation — Cloud Security & DevSecOps Engineering
- Continuous Delivery Versus Continuous Deployment — DevOps & Site Reliability Engineering
- Deployment Strategies: Canary Releases — DevOps & Site Reliability Engineering
- Decoupling Release From Deployment: Feature Flags — DevOps & Site Reliability Engineering
- Continuous Delivery: Blue-Green & Canary Pipelines — Production Full-Stack Engineering
- Least Privilege In IAM And RBAC — Application Security & Threat Modeling for Engineers
- The Environment Variable Trap — Application Security & Threat Modeling for Engineers
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture
- Worked Design Three: Scaling The Metrics Pipeline — Scalable System Design & Distributed Architecture
- The Test Pyramid And Where It Is Wrong — Automated Testing, TDD & Quality Engineering
References
- Google SRE workbook: Canarying releases
- AWS Well-Architected Reliability pillar: use fault isolation to protect your workload
- AWS Well-Architected: reduce blast radius
- Amazon Builders' Library: Workload isolation using shuffle-sharding
- Martin Fowler: TestPyramid