Edge App

Mellerikat Edge App is a deployment solution that enables fast and reliable distribution of AI models across various edge environments. AI models trained via AI Conductor can be effortlessly deployed to Edge App using Edge Conductor with a single click, making them immediately usable in the field.

Edge AI

AI Model Deployment Across Diverse Environments

By installing Edge App on various operational environments—such as PCs, servers, cloud platforms, and IoT devices—AI models can be deployed with ease. With support for diverse installation environments, Edge App enables users to implement AI solutions flexibly and efficiently, ensuring optimal performance across different use cases.

Real-time AI Inference Monitoring

Real-Time Inference Result Monitoring

Using AI models deployed through Edge App, real-time inference can be performed, and the results can be monitored live via Edge Viewer. Edge App delivers high-performance and accurate inference results, while Edge Viewer enables real-time monitoring of those results—supporting fast and efficient data analysis and decision-making.

Seamless Updates

Flexible AI Model Replacement Scenarios

The AI Solution Pipeline and AI Model running on Edge App can be flexibly replaced. If the performance of the deployed AI model degrades or a new model is needed, users can easily replace both the AI Solution and the AI Model through the Edge Conductor connected to Edge App. This flexible model replacement capability ensures that Edge App consistently maintains optimal performance, even across diverse operational environments.

Cost-Effective

Cost-Effective Operational Architecture

Edge App can be turned on or off based on user needs. When the AI model is not in use in the environment where Edge App is installed, the app can be turned off to conserve system resources. When restarted, Edge App retains the previously configured AI model and environment settings, allowing users to resume AI model usage at any time without reconfiguration. In on-premise environments, users can control power states via Edge Viewer, while in cloud environments, scheduling can be managed through the Edge Conductor UI. These features provide the flexibility to efficiently manage AI models and conserve resources in any environment where Edge App is installed.

Architecture

Micro Service Architecture

Edge App is designed based on a microservice architecture, where the control unit, data input/output unit, and inference unit operate independently from one another. In particular, the AI Solution Pipeline and AI Model, which comprise the inference unit, can be flexibly replaced—allowing users to easily update or switch AI solutions as needed. Moreover, this robust fault isolation structure ensures that even if inference fails or there are issues with data input/output, other components continue to operate reliably—providing stable and uninterrupted service.