Glossary

Every Antithesis-specific term we use in the docs is collected and alphabetized here. We also have a distributed systems reliability glossary that covers terms used in this space more broadly.

Assertion: A way to express properties programmatically so Antithesis can test them. Unlike the assertions you know from unit testing, they don’t stop execution when they fail. Covered in depth here.
Assertion catalog: Antithesis registers every assertion added in the system under test and the workload in an assertion catalog, which it uses to evaluate the corresponding properties.
Autonomous testing: Using software to generate test cases, instead of a human developer. Covered in depth here.
Branch: A portion of a timeline downstream of a particular fork in the multiverse map. Generally used in the sense of “a particular event created a new branch.” Covered in depth here.
Default properties: Antithesis comes with a built-in set of properties that should hold in all systems.
Deterministic simulation testing: Testing a system by running it in an environment simulating some or all layers of the testing stack. Sources of non-determinism like clocks, thread interleaving, and system-provided sources of randomness (among others), are made deterministic in the test environment. Covered in depth here.
Event: An occurrence within the simulation. Commonly, a log message from your system, an SDK assertion, or a fault event that Antithesis triggered. Every event occurs at a unique moment.
Example-based testing: Tests which dictate the inputs to be applied to the system under test, and expect a specific result. This is the approach taken by most software testing.
Moment: Used as in common English, a unit of time on a timeline. Each moment occurs on a particular timeline, and there are some moments where no events occur.
Multiverse: The collection of timelines explored during a single test run. You can think of each test run as producing its own multiverse.
Property: An invariant. A promised or expected behavior of a system that we want to test. Covered in depth here.
Property-based testing: Testing that uses computer-generated random inputs to check that a function or system behaves as expected under a wide range of conditions. Covered in depth here.
Setup: The initial phase of the test run when your system under test and your workload are deployed and made ready for exploration.
System under test (SUT): The system you’re testing.
Test environment: AKA “the Antithesis environment” or just “the environment.” The deterministic simulation environment where the SUT is deployed and run during the test.
Test run: AKA “run”. A single testing session in Antithesis, during which multiple timelines are explored. A single test run generates a single multiverse.
Timeline: One complete execution history, from start to end, that your system took during the test run. Exploring your software explains this concept in detail.
Triage report: A report consolidating the results at the end of a test run.
Virtual time: The simulated clock used to order events during a test run.
Workload: Code that drives activity in your system during an test run. Antithesis observes the system behavior to detect failures.