Glossary

Docker image that can be mounted inside projects using one of the integrated technologies. Persistent storage spaces can be assigned to an app to keep data in a defined volume path when the app is stopped.

An open source tools for Kubernetes to run workflows. An open source container-native workflow engine for orchestrating parallel jobs on Kubernetes. For more information, see the Argo documentation.

Rights granted to groups through the Security module. These rights include access to specific features of the app, access to datasets, and project editing rights.

A Saagie component to import files for jobs and apps. Files are managed on a Kubernetes platform and accessible through that platform. Saagie users working with Kubernetes clusters can use the exposed Conf’o’rama API.

A collection of data that is accessible through the Governance data portal.

Set of rules that define which groups can access which datasets in Saagie.

Login information stored in projects. It allows you to pull Docker images from private registries when creating apps.

An entity that groups datasets in the Governance module. Domains are organized in a flexible way to match the data organization of the company. For example, by geography, branch, or team.

Dynamic values that can be referenced in projects, pipelines, apps, and also jobs. They limit the need to modify a value in the execution code each time this value changes. They can be created at platform (global), project, pipeline, and app levels. This ensures proper control over variable access and use.

An external connection is a connection parameter specific to an external technology provider. It is used when creating external jobs in projects. It includes at least the remote host information and authentication requirements used to connect to the external job technology.

Multi-platform log processor and forwarder that collects data and logs from different sources, then unifies and sends them to multiple destinations. For more information, see the Fluent Bit documentation.

RedHat tool used for Identity Management. For more information, see FreeIPA.

Entity in which a selection of users is placed. Groups are granted authorizations in the Projects and Governance modules. When an administrator grants authorizations to a group, each user in that group receives those authorizations.

Helm Charts help you define, install, and upgrade Kubernetes applications. For more information, see the Helm documentation.

A single run of a job or pipeline in a project. Execution information and job logs of all instances are saved on your platform.

Computation task performed inside projects on Saagie. It uses one of the embedded or external technologies. Jobs run through a command line and can be launched individually (in both manual and scheduled modes) or as part of a data pipeline.

Also called Apache Kafka, it is a distributed streaming platform used to facilitate communication between systems and applications. It must be used with ZooKeeper. In Saagie, Kafka is used to manage communication between Saagie components. For more information, see the Kafka documentation.

Security-oriented federation and identity provider (IdP) server. In Saagie, Keycloak is used with LDAP to get security tokens. For more information, see the Keycloak documentation.

Legacy user interface of Saagie. It allows the creation of extraction, processing, and smart app jobs, as well as access to data lake and data mart services. In the Manager module, all jobs are listed in a single menu, while in the Projects module, they are organized by project to sort them more easily.

Tool used in Saagie legacy orchestration. Chronos schedules jobs in Mesos. Marathon ensures that jobs restart automatically if a job fails in Mesos.

Object storage server. It is compatible with Amazon S3 cloud storage service. In Saagie, MinIO is installed in the Kubernetes cluster to store files. For more information, see the MinIO documentation.

Document-oriented NoSQL database that uses JSON-like documents with optional schemas. Queries are made in JavaScript. For more information, see the MongoDB documentation.

Relational database management system. For more information, see the MySQL documentation.

The package is either a file, or a collection of files in a .zip file. Compatible file types change according to the technology selected, and are listed when uploading a package.

End-to-end process encompassing multiple jobs that are connected and executed within projects. A pipeline can include advanced orchestration logic, such as conditions or parallel job executions.

Kubernetes controller created by Saagie to manage CRD platforms in a Kubernetes cluster.

Relational database management system. For more information, see the PostgreSQL documentation.

Kubernetes controller created by Saagie to manage CRD projects in a Kubernetes cluster.

Component created by Saagie to manage projects and their elements, such as jobs, apps, or pipelines. Projects-and-jobs' API is exposed and can be used by all Saagie users.

Secures and manages security isolation for a given Saagie customer. In Saagie, it is a string used in the application URL and request headers.

Component created by Saagie to store Docker credentials used for private images in jobs and apps.

A storage space is persistent. It can be assigned to an app to keep data in a defined volume path.

Single Saagie account created on one or more platforms. Authorizations are granted to users by administrators.

A single iteration of a job, pipeline, or app. Each new upgrade is stored as a version, so you can roll back to previous iterations and keep track of the changes that have been made.

Kubernetes’ resource created by Argo that facilitates the creation and management of workflows. In Saagie, we use it to operate pipelines and jobs.

Centralized service used to maintain configuration information and provide distributed synchronization and group services. In Saagie, ZooKeeper is mainly used to maintain and operate a Kafka cluster. For more information, see the ZooKeeper documentation.