Article

Run Anywhere On-Premise or in the Cloud

APM Studio’s architecture is engineered for flexibility and scalability, allowing you to deploy diagnostic and prognostic models on real-time data—whether you’re working on-chip, at the edge, or in the cloud. Built with containerization and platform-agnostic design in mind, APM Studio enables consistent, high-performance analytics across diverse environments, from compact ARM-based devices to full-scale server clusters.

A digital illustration of cloud computing with interconnected icons representing data, analytics, communication, and IoT devices, overlaid on a modern cityscape with skyscrapers.

APM Studio’s adaptable, containerized architecture supports everything from on-chip deployments to full-scale cloud clusters—enabling real-time data processing wherever it’s needed. Whether you’re working with ARM-based devices at the edge or orchestrating Kubernetes environments across public and private clouds, APM Studio ensures the same robust analytics platform follows your operations, no matter where they live.

A deployment architecture diagram for APM Studio, illustrating scalability from single devices to cloud infrastructure. It includes deployment options on devices, edge/PLC, on-premise, and cloud environments like DigitalOcean, AWS, and Azure. Docker and Kubernetes are highlighted for containerization and scalability, with logos of hardware partners such as WAGO, Phoenix Contact, Hilscher, and Balluff.

Versatile Architectures (ARM32/ARM64/x86)

APM Studio supports a wide range of hardware, from small embedded devices to traditional x86/AMD64 servers. This means you can run real-time analytics:

  • On Chip – Leverage ARM-based systems for on-device intelligence, minimizing latency and reducing reliance on external connectivity.
  • Edge/PLC – Deploy directly on PLCs from vendors like WAGO, Phoenix Contact, or Hilscher to bring analytics closer to your industrial processes.
  • On-Premises – Install on x86 or ARM servers within your facility for secure, locally managed data processing.

Containerization with Docker

By packaging APM Studio in Docker containers, you can streamline deployment across multiple targets:

  • Consistent Environment – Eliminate version conflicts by encapsulating the entire software stack.
  • Easy Updates – Pull the latest container to upgrade your analytics capabilities without worrying about dependencies or OS variations.
  • Rapid Scalability – Spin up additional containers to handle increased data loads or new assets, either on-premise or in the cloud.

One Code Base, Multiple Deployments

APM Studio uses a single codebase for all deployment types—on-chip, edge, or cloud. This ensures:

  • Unified Features – The same ML libraries, Bowtie models, or integration connectors run consistently, regardless of location.
  • Simplified Maintenance – Write your analytics rules or custom scripts once and deploy them wherever you need—no separate code branches or device-specific adaptations.
  • Streamlined Updates – Patches, feature enhancements, and new models become available instantly across all deployment footprints.

Cloud-Agnostic or Private Infrastructure

Whether you opt for UReason’s managed cloud, public clouds (DigitalOcean, AWS, Azure), or private, on-premise data centers, APM Studio adapts to your existing IT stack:

  • Kubernetes Support – Orchestrate large-scale deployments with automated load balancing, rolling updates, and high availability.
  • Hybrid Flexibility – Combine on-premises installations with public cloud expansions for burst capacity or multi-site strategies.
  • Security & Compliance – Keep sensitive data on-prem if needed, while still benefiting from cloud-based analytics scaling.

Benefits of Run-Anywhere Deployments

  1. Low Latency & High Reliability
    Deploy on embedded or edge devices to process data in real time, ensuring immediate feedback and localized decision-making.
  2. Optimized Resource Usage
    Choose the right environment (edge or cloud) based on throughput requirements, cost considerations, and connectivity constraints.
  3. Scalable to 1,000,000+ Devices
    From a single sensor to an enterprise-wide fleet, APM Studio’s architecture accommodates any scale—spinning up containers or Kubernetes nodes as needed.
  4. Consistent Analytics Pipeline
    Keep a uniform approach to condition monitoring, predictive maintenance, and AI/ML across devices and sites—no fragmented solutions.
  5. Reduced Operational Complexity
    A single codebase and containerized approach simplify updates, patches, and new feature rollouts—both on-prem and in the cloud.

Example Deployment Scenarios

  • On a Compact ARM Device:
    A pump manufacturer embeds APM Studio analytics (deterministic rules + AI) inside a low-power ARM chip for local fault detection—no cloud connection required.
  • At the Edge with PLC Integration:
    An industrial plant running PLCs from Phoenix Contact uses Dockerized APM Studio instances at each line to handle high-volume sensor data locally, then forwards critical insights to a central control room.
  • Fully Cloud-Based:
    A multinational organization deploys APM Studio in Kubernetes clusters on AWS for a global asset base, offering 24/7 scaling and resilience.

Whether you need embedded analytics at the device level or enterprise-scale computing on Kubernetes, APM Studio’s containerized, platform-agnostic architecture ensures you can run anywhere—quickly, reliably, and with minimal operational overhead.

Ready to take the first step?

Book a call with Artur Loorpuu, Senior Solutions Engineer in Digitalization, who specializes in turning industrial challenges into practical digital solutions. With deep expertise in digitalization and process optimization, Artur collaborates with clients in the process industry to reduce costs, enhance efficiency, and drive innovation.

Let’s explore how we can support your goals!

Related Articles