Flexible and Scalable Platform Litmus Edge collects, normalizes, analyzes and integrates real-time data
from any source to provide a complete data picture across the organization with one single solution.

• Deploy securely and locally at the edge with no internet connection required
• Install as an OS on any gateway, VM or server at the edge
• Deploy quickly on a pre-installed device or via USB
• Access an intuitive web interface via web browser or terminal user interface (TUI)
• Securely manage any number of assets at any number of sites
• Access Litmus Edge anywhere in the world and diagnose issues remotely

Device Connectivity Quickly connect to any industrial asset – PLC, DCS, SCADA, Historian, sensor or ERP – with pre-loaded drivers and no programming required.

• Scan the network to seamlessly add industrial assets without coding
• Access 250+ legacy industrial systems and protocols out-of-the-box
• Use the DeviceHub interface to add, modify, refresh, start, stop or remove a devi ce
• Define how to connect and collect data from any device on the net work
• Use a drag-and-drop flows editor to test device connectivity and customize workflows

Data Collection Collect and normalize hundreds of custom data points from any number of assets into one standard format for consumption by any application.

• Store normalized data in a scalable and secure time series database
• Index data to be utilized for terabytes of storage
• Use an optimized version of influxdb for all data storage
• Publish data to a local message broker for immediate consumption
• Access native data by SDKs and non-native REST API
• Integrate with any enterprise grade cold storage
• Every data point has pre-analyzed cubes for device management, alerts and analytics

Real-time Analytics Monitor real-time asset data, set up alerts and utilize ready analytics based on common KPIs such as uptime, downtime, anomaly detection and more.

• Use ready analytics to dramatically reduce manual setup and configuration time
• Configure KPIs including OEE, uptime, downtime and more with no coding
• Configure time series data analytics by average, maximum and minimum
• Perform statistical and analytical queries on the live data
• Define workflows with a drag-and-drop editor for simple data manipulation and visualization
• Utilize Grafana open source-based dashboards
• Create visualizations, BI dashboards and custom SQL-scripted analytics in a few clicks

Application Marketplace Enable one-click application orchestration and deploy docker container-based applications from a public or private application marketplace.

• Access the Litmus Edge Marketplace, a local application repository for launching applications on demand to enable edge-level analytics
• Utilize a default set of 45+ applications in the Public Marketplace
• Add a Private Marketplace to leverage existing custom applications
• Add proprietary docker container-based applications to the Marketplace
• Deploy docker applications to one or many devices with one click
• Zero-touch provisioning, mass management and application orchestration
• Perform application orchestration and lifecycle management (in docker) from a central location

Data Integration Immediately feed valuable and ready-to-use data to any cloud or enterprise application to achieve a complete data picture from OT to IT.

• Easily integrate to the cloud with pre-built connectors for data visualization and device management
• Feed collected data into Big Data implementations with native Kafka and database interface
• MQTT for device-to-Litmus Edge data collection and pre-processing
• REST API integration for workflows

Machine Learning Runtime Feed machine learning models with normalized data and complete the feedback loop by running the new models at the asset for continuous optimization.

• Run machine learning processes inside Litmus Edge ready analytics
• Utilize availacle models for precliction, classifiction and anomaly cletection
• Save models from TensorFlow and upload them to Limus Edge analytics
• Access pre-built, easy to configure connectors to Cloudera, Azure, Oden and others for rapid machine leaming deployments
• Feed machine learning models with normalized data
• Enable edge-sicle models to ingest data, form local devices and act based on traning received from cloud-based platforms
• Run new models at the asset to deliver corrective actions in real-time