Executionware: Empowering Cloud and Edge Deployment in NebulOuS

In today’s dynamic cloud-edge continuum, seamless application deployment across diverse environments is essential. As part of the EU’s NebulOuS project, Activeeon advances resource provisioning and execution management technologies for cloud and edge environments. This overview highlights how Executionware enables cloud applications to efficiently manage and scale across heterogeneous infrastructure. 

The Role of Execution in Cloud Applications

Execution in cloud computing refers to deploying, operating, and managing workloads across various cloud resources. Effective execution automates two key tasks: 

  • Resource provisioning: Allocating computing resources (e.g., virtual machines, containers, or physical servers) to meet application needs.
  • Environment setup: Configuring infrastructure by installing dependencies and networks to ensure smooth operation.

For distributed applications, especially those with complex workflows, the execution layer orchestrates resources across private, public, and edge clouds. Without automated execution management, developers face the tedious task of manually managing setup, scaling, and maintenance, which grows more challenging as applications expand. Activeeon simplifies these tasks, automating execution and enabling seamless deployment across diverse environments. 

Bridging NebulOuS Upperware and Infrastructure

In the NebulOuS architecture, Executionware automates the deployment and orchestration of cloud and edge applications. It comprises two key components: the Execution Adapter and the Deployment Manager, which together manage workloads across cloud, edge, and on-premise environments. 

  • Execution Adapter: Powered by ProActive Workflows & Scheduling technology, this component connects abstract deployment descriptions with actual cloud resources. It directly interacts with cloud provider APIs to provision resources and deploy applications across public, private, or edge environments, ensuring that applications scale smoothly across platforms by abstracting cloud complexity.
  • Deployment Manager: The Scheduling Abstraction Layer (SAL) collaborates with the Execution Adapter to manage resource allocation and deployment. SAL handles tasks like registering/deregistering cloud providers and edge devices, selecting compute resources (e.g., location, CPU, memory), and supporting dynamic reconfiguration for optimal deployment. It interfaces with key NebulOuS components like the GUI, Optimizer and Fog/Edge Resources Manager, efficiently allocating resources. Additionally, SAL translates high-level application requirements into executable tasks and automates the installation of essential software (e.g., Kubernetes, Docker, Kubavela, Knative), along with NebulOuS components such as the ONM, EMS and Optimization Solver.

Unlocking the Cloud-Edge Continuum

By integrating the Execution Adapter and Deployment Manager (SAL), Activeeon’s Executionware automates resource provisioning, environment setup, and application execution across cloud and edge environments. This significantly reduces the complexity of managing distributed applications, enabling organizations to fully leverage the cloud-edge continuum. Activeeon’s automation allows businesses to focus on their core operations, confident that their infrastructure is efficiently managed.

ANKICA BARIŠIĆ

ANKICA BARIŠIĆ

R&D Engineer focused on optimising Cloud and Edge infrastructures at Activeeon

Ankica Barišić holds a Ph.D. in Computer Science from NOVA University Lisbon, with postdoctoral research experience at institutes across Portugal, Croatia, and France. Her expertise centers on Software and Systems Engineering, with a focus on Embedded and Cyber-Physical Systems. Specializing in Model-Driven Engineering, she develops tailored modeling techniques that enhance usability, cross-domain collaboration, and sustainable system development. Bridging academic and industry settings, her work translates theory into practical, sustainable solutions for complex systems engineering.

Share this article

Post
Share

SUGGESTED ARTICLES