9.2. Clara Deploy SDK Pipeline Glossary

See Docker for additional information.

See Docker for additional information.

In mathematics, particularly graph theory, and computer science, a directed acyclic graph (DAG /ˈdæɡ/ ), is a finite directed graph with no directed cycles. That is, it consists of finitely many vertices and edges (also called arcs), with each edge directed from one vertex to another, such that there is no way to start at any vertex v and follow a consistently-directed sequence of edges that eventually loops back to v again. Equivalently, a DAG is a directed graph that has a topological ordering, a sequence of the vertices such that every edge is directed from earlier to later in the sequence.

See Wikipedia for additional information.

Per pipeline job resource management container. Payloads are defined as a set of input/output directories intended to be used by a pipeline’s operators. Payloads are allocated, managed, and deallocated by CLARA Platform.

Configurable data processing model containing one or more interconnected stages, or operators. Pipelines must be designed as a directed acyclic graph) where a pipeline’s operators are vertices and interconnections are edges.

See Pipelines for additional information.

Executable code which performs an operator‘s data processing function. Each operator is defined by the execution of a single pipeline application.

An instanced execution of a pipeline.

Subtask of a pipeline, intended to operate on a set of inputs and produce a set of outputs.

See Operators for additional information.

Long running container application which can support multiple operators in parallel. A Service’s life-time is at least as long as the pipeline’s in which the services has been requested.

See Services for additional information.

© Copyright 2018-2020, NVIDIA Corporation. All rights reserved.. Last updated on Feb 1, 2023.