Definitions

📘

This page gives detailed definitions of various terms found throughout the learn docs as well as those that will be seen through the use of the platform

API

An API (Application Programming Interface) is a set of protocols, tools, and standards for building software applications. It allows different software systems to communicate and exchange data with each other.

CLI (Command Line Interface)

A Command Line Interface (CLI) is a text-based user interface (UI) used to run programs, manage computer files, and interact with the computer. CLIs accept text as input commands and are entered using a keyboard; the commands invoked at the command prompt are then run by your computer.

Marketplace

The Eratos Marketplace is where you can find and browse Products. Products refer to datasets, workflows, and solutions published by members of the community, along with any important contextual information pertaining to the Product.

You can also package up and publish your own work to the Marketplace as a Product, making your IP available to other members of the community, for free or as a paid Product.

Members of the community can have their own dedicated creator page that is associated with the solutions and datasets they have published. This enables members to discover and collaborate with each other, to solve complex problems.

Contextual Links

Similar to the semantic web, Eratos creates links between resources. This linkage is important as it gives context to the data, model, and output, resulting in the ability to track and trace upstream and downstream actions and connections in an easy and intuitive manner. This effectively creates knowledge graphs that can be used to track things such as data provenance, detailed permissions, auditing of data transformations, and data ownership.

Datasets

Eratos is a distributed data infrastructure that provides easy, uniform access to disparate and dispersed data. Petabytes of data are available through Eratos, all with strong findability, data provenance, and access control.

An Eratos dataset is a collection of related data that has been indexed. Eratos makes it easy to find and use datasets without needing to know how the underlying data are stored, structured, or formatted.

Blocks provide a convenient way to create Eratos datasets, either by uploading the data directly or by describing its location and the protocols for accessing it (e.g. via a THREDDS Data Server or an Eratos Gateway Node).

Distributed Architecture

The Platform has been designed to make the toughest data on the planet accessible - geospatial data. Being optimized for highly fragmented, disparate, and variable datasets, Eratos can very easily manage any data type. To make this possible, the platform was architected to operate in a distributed manner, meaning data can reside in its original location instead of requiring users to centralize it. For programs such as the FCDI, there are over 10 petabytes of data located at 8 locations around Australia. Centralizing this data is difficult and costly due to the needs of the various data owners. The Platform has made it possible to access all of this data.

Keys

The combination of your Key ID and Key Secret is required to access the parts of Eratos that you have permission for. They provide the user with the ability to authenticate themselves and access the system.

Licensing

Licensing terms and conditions for Eratos

Eratos Marketplace License Terms and Conditions

Eratos Privacy Policy

Eratos Subscription Terms and Conditions

Products made available within the Eratos Marketplace, or shared within private Eratos Spaces can select from a range of licences. Details of each licence available for Eratos users to apply are listed below.

Model

A Model is a well organised tool or script that processes and input using an operator and produces an output. The Model framework enables models to be reused easily on different data sources. Multiple Models can be composed together to quickly create flexible and reliable workflows. Models are comprised of 3 Resources, which are defined in metadata JSON files.

Operator: calculate the average mean temperature at a specific location over time.
Inputs: location, start and end dates, temperature.
Output: average mean temperature.

Nodes

To manage a distributed environment with data and models stored in multiple locations, Eratos runs what are called Primary and Gateway nodes. Gateway Nodes run where the data is stored in order to manage datasets, including storage, access, and translation. A small protocol runs at the data location to index the data into metadata, create users and permissions, and access the Primary Node. The Primary Node aggregates Gateway Nodes to effectively provide an organizational view and administration layer. Nodes may be deployed by a user privately or users may make use of the Eratos managed gateway nodes.

Operator

An Operator is a generic function or piece of code that requires an input and provides an output. When combined with a metadata JSON file, it can be hosted on Eratos. Operators are key components of Eratos Models. Note that many operators can be used in conjunction when building a model.

Product

An Eratos Product refers to a dataset, model, or solution that is available within the Community or Space. An Eratos Product uses the technical concept of a "block" to represent the Product. Blocks are data or code that has been packaged in an Eratos wrapper making them easily combined to create custom solutions. They may be likened to building blocks in that they are modular, composable units of data and/or code.

Resources

An Eratos Resource is a semantic object, a collection of metadata that describes attributes and links relating to a distinct entity. Resources can represent many different types of entities like a: dataset, operator, API, physical object or place, concept, collection, etc.

Resources give the data in Eratos context and meaning, which is needed to make data FAIR (Findable, Interoperable, Accessible, and Reusable).

Every model, dataset, and product in Eratos has a Resource that describes it.

SDK (Software Development Kit)

A software development kit is a collection of software development tools in one installable package.

Senaps

Eratos integrates a technology developed by the Australian Government’s Commonwealth Scientific and Industrial Research Organization (CSIRO) called Senaps. The technology is exclusive to Eratos and is an enterprise IoT workflow and model orchestration engine.

It allows Eratos to connect to any IoT or embedded network, extract near-real-time data, and transform that data in-flow for analysis and reporting.

Mixed with Eratos’ ability to curate time-series data, Senaps is a powerful monitoring tool, with an ability to reuse real-time data for scenario and longitudinal analysis.

Solution

A Solution "block" is a piece of code that performs a function or set of functions to solve a specific problem. It can be a stand-alone piece of code that requires the input of data, or it can include a dataset. Because they are packaged up in the form of a modular and composable "block", provided they are compatible, multiple datasets and solutions can be combined to form a single Solution.

Spaces

Spaces allow users to manage, innovate, and collaborate within a dedicated space.
Spaces are collections of members, resources, projects, and workflows.

You can create Spaces for different individuals, teams, departments, or organization units, depending on the required context. Owners and admins manage members, permissions, and settings, and assign dedicated admins to their Space.

Spaces: Access

The Eratos Workspace is where you can privately access datasets, workflows, and solutions published by members of the Workspace, along with any important contextual information. All assets inside the workspace can only be accessed by members in the workspace.

Spaces: Collaborate

Workspaces allow you to manage, innovate, and collaborate within a dedicated space. New members and their level of permissions can be invited and managed by the owner and administrators of the workspace.

Spaces: Share

You can also package up and privately share your own work to the Workspace as a Product, making them freely available or commercializing them. Members of the Workspace can then use them in their own solutions.

Standardised Data

Eratos uses a concept called a Type Interface. This provides the automatic translation of one type of data into another type so that it can be easily ingested into modeling, visualization, or other systems. For example, through a gridded data connector, a user can extract climate data into a CSV if they so wish, without the knowledge that the data is stored in a gridded data file. This is a powerful feature that significantly reduces the complexity of working with disparate data at scale.

Studio

Studio is a no-code front-end application on the Eratos Platform where users can run Operators to produce an output. Users specify the relevant inputs on the front end and the Operator is run without the need to utilize the technical SDK.

Terminal

Terminal is the console for Mac and Linux or the command prompt for Windows. You can check that the CLI is installed, and see the help information using eratos --help

User

An Eratos User is defined as anyone with an Eratos Platform registered account, allowing authorized access to the Platform.

User Roles

Owner - The owner of a space within the Eratos Platform. The owner is ultimately responsible for billable usage related to the Space.

Admin - The administrator of a space. Able to invite other users, manage user access permissions, invite additional users, and add Products to the Space.

Contributor - The contributor is able to access and run workflows & run operators. They are also able to publish their results onto the workspace.

Reader - The reader has read access to the space but is unable to create, administrate other users, run workflows, or run operators.

Variable Extraction

A powerful feature in Eratos is an API that allows users to extract subsets of data from data files. An example of how this is being used today is how climate scientists are working with the FCDI (Federated Climate Data Initiative) data.

Imagine tens of thousands of files located across 8 major organizations. The files can contain 140 years worth of 200+ variables (e.g. wind, temperature, rainfall, humidity), in time steps of one minute across all of Australia - each file being about 80MB. Climate scientists are extracting data on specific variables, decades, locations, and time steps in minutes. The result is model creation and analysis that has traditionally taken months to complete being completed in hours or days.

Viewer

The Eratos Viewer is a Grafana plugin that allows for data visualization of pre-defined templated solutions.

Workflow

A Workflow is an object that targets a particular set of components (e.g. datasets, operators) that have been configured in a specific sequence to achieve a desired outcome/output. Once set up, components may be re-used and configured multiple times across different workflows, allowing for economies of scale and standardized datasets.