Frequently asked questions
What is autonomous infrastructure remediation?
Autonomous infrastructure remediation is the ability to detect that a system has diverged from its intended state and correct it without a person executing the change. In practice there are two levels. The first is reconciliation: the intended state is declared, the difference is detected, and the environment is brought back in line. The second is deciding what to do about a problem that is not a simple divergence, and there the decision has to come from somewhere: a rule, an operator, or an external analysis system. The two are often described together, which hides the fact that only the first is a closed loop.
What is the difference between observability and remediation?
Observability helps organizations detect and understand infrastructure issues by collecting metrics, logs, and traces. Remediation goes one step further by executing the actions required to resolve those issues. Together, they enable faster and more reliable operational responses.
Why is deterministic automation important for infrastructure operations?
Deterministic automation executes only predefined and approved operations based on explicit rules rather than probabilistic decisions. This makes automated remediation predictable, explainable, and fully auditable, which is essential in regulated and mission critical environments.
How does Fractal Cloud execute automated remediation safely?
Because the action space is defined in advance. A Fractal declares which lifecycle operations are allowed on a system, and nothing outside that set can be executed, whoever or whatever requests it. Fractal Cloud continuously reconciles each Live System against its declared state. When a remediation decision comes from outside, for example from a monitoring or analysis platform, Fractal Cloud acts as the actuator and executes it inside the same boundaries. Every action is recorded against the Fractal that authorized it.
What is a Digital Twin of the Organization?
A Digital Twin of the Organization (DTO) is a live model of an IT estate that represents not only the individual systems but the relationships between them. Traditional monitoring tells you that a component is failing. A DTO tells you what else depends on that component, so an event can be assessed by impact rather than by alert volume. It is a category of observability and analysis product rather than an infrastructure platform: the model produces a decision, and something else has to carry it out.
