Here are some definitions of terms commonly used throughout this site.
Class of problems where the goal is to determine behavior that is sufficiently different from the nominal (or normal) behavior patterns.
Application programming interface.
Allows data scientists to publish machine learning models into the Arundo Fabric cloud environment with a single command. Arundo Composer is made up of two components:
Arundo Composer CLI
Command-line interface that interacts with your data science environment. This powerful command-line tool gives you the ability to:
- Deploy models for testing
- Publish models into Arundo Fabric
Arundo Composer Runtime
Package of modules that builds REST endpoints for a model. This gives you the ability to:
- Test the model in a model application (web browser-based application)
- Interact with the model using third-party applications
aka Edge, Edge Agent
IoT gateway that enables cloud-based analytics for industrial assets. Arundo Edge connects to industrial control systems and retrieves data from sensors on industrial assets. Arundo Edge then streams the sensor data to the cloud, where operations personnel can run the data through machine learning models and interpret model results on dashboards.
Application suite that provides software to enable enterprise-scale machine learning and advanced analytics applications for industrial companies. Arundo Enterprise transforms businesses through predictive analytics, deeper understanding of physical operations, and data-driven software-as-a-service (SaaS) applications.
Arundo Enterprise features the following components:
- Arundo Fabric
- Arundo Composer
- Arundo Edge
Cloud environment at the core of the Arundo Enterprise suite. Arundo Fabric collects streaming data from edge devices, connects machine learning models, establishes data pipelines, and displays visualizations of data in extended applications.
Arundo REST API¶
Features a set of HTTP methods that you can use to interact with Arundo Enterprise. This restful API gives you the ability to interact with data and develop integrations with third-party applications.
Instance of Arundo Enterprise unique to your organization. All Arundo tenants run in the Microsoft Azure cloud with their own dedicated resources.
Equipment, machinery, vehicle, or vessel operated during an industrial process.
Command-line interface. Application that operates using text commands. Arundo Enterprise features a command-line interface called Arundo Composer CLI, which gives you the ability to test and publish machine learning models to Arundo Fabric.
Command-line package and environment manager for Python. Conda is a powerful package manager and environment manager that you use with command line commands at the Anaconda Prompt for Windows, or in a Terminal window for macOS or Linux.
Device installed on-site ("on the edge") next to an industrial asset. In the Arundo world, edge devices are where you install Arundo Edge to collect sensor data.
Location of the data center where your Arundo Enterprise instance is located. There are currently two environments:
Shortcut to an application that displays on the Home page in Arundo Fabric.
Industrial Internet of Things. Use of IoT technologies to enhance manufacturing and industrial processes.
Internet of Things. Network of devices connected to the internet that collect and share data.
Device that connects non-IP sensors to the internet. You can set up Arundo Edge on a device to turn it into an IoT gateway.
Class of algorithms that learn representations of data sets. By passing in data during the training phase, the internal parameters of the algorithm are optimized to provide the best predictive performance.
Cloud computing platform on which Arundo Enterprise operates.
Serial communication protocol for use with programmable logic controllers (PLCs).
aka machine learning model, ML model
Contains algorithms that analyze what happened in the past to help us do better in the future.
Web browser-based app you can use to interact with a machine learning model. This lets you test a model locally in a controlled environment.
REST endpoint created by publishing a model in Arundo Composer. A model endpoint contains a model that takes inputs, executes logic, and generates an output. You can interact with a model endpoint in a model application.
Stores the files necessary to build, test, and publish a model.
On the Edge¶
Describes applications and components that run on-site next to the industrial assets from which they gather data.
Interoperability standard for the secure and reliable exchange of data in the industrial automation space and in other industries. OPC is platform independent and ensures the seamless flow of information among devices from multiple vendors.
aka OPC Unified Architecture
Machine-to-machine communication protocol for industrial automation developed by the OPC Foundation.
Sends data through a machine learning model and outputs results for analysis. There are two types of pipeline:
|Streaming||* Receives streaming data from Arundo Edge
* Runs continuously to output real-time data
|Batch||* Receives data from Arundo Fabric warm storage or a comma-separated values (CSV) file
* Runs on-demand or at a preset interval to output data for a defined period
aka Programmable Logic Controller
Industrial computer control system that continuously monitors the state of input devices and makes decisions based upon a custom program to control the state of output devices.
Programming language that you can use to develop machine learning models.
You must use Python in the app.py files for models.
Device that measures physical properties of an industrial asset and records these measurements for analysis.
Label for data. In Arundo Enterprise, a tag can be one of the following:
- Input or output for a machine learning model in a pipeline
- Data source for a panel on a dashboard
Time Series Data¶
Series of data points indexed, listed, or graphed in time order. Usually, a time series is a sequence of data points collected at successive equally spaced points in time.