SODALITE
Research Area:
Research Lines:
Research Lines:
Horizon 2020
DEIB Role: Partner
Start date: 2019-02-01
Length: 36 months
Project abstract
SODALITE - Software Defined Application Infrastructures Management and Engineering
SODALITE will provide application developers and infrastructure operators with tools that abstract their application and infrastructure requirements to enable simpler and faster development, deployment, operation, and execution of heterogeneous applications reflecting diverse circumstances over heterogeneous, software-defined, high-performance, cloud infrastructures, with a particular focus on performance, quality, manageability, and reliability. In doing so, it will produce several tangible results:
SODALITE will provide application developers and infrastructure operators with tools that abstract their application and infrastructure requirements to enable simpler and faster development, deployment, operation, and execution of heterogeneous applications reflecting diverse circumstances over heterogeneous, software-defined, high-performance, cloud infrastructures, with a particular focus on performance, quality, manageability, and reliability. In doing so, it will produce several tangible results:
- a pattern-based abstraction library, including application, infrastructure and an absolute novum, performance abstractions;
- a design and programming model for full-stack application and infrastructure descriptions, using abstraction library;
- a deployment framework, enabling static optimisation of the so-abstracted applications onto specific infrastructures;
- automated run-time optimisation and management of so-deployed applications.
- increasing design and runtime effectiveness of software-defined infrastructures, to ensure high-performance execution over dynamic heterogeneous execution environments;
- increasing simplicity of modelling applications and infrastructures, to improve manageability, collaboration, and time to market.
Project results
SODALITE has developed a robust and powerful end-to-end toolkit to abstract application and infrastructure requirements with a particular focus on performance, quality, and reliability. Following a layered architecture approach, SODALITE tools provide specific advanced functionalities:
- SmartIDE: Shareable knowledge base and a semantic reasoners API which provides inputs and a DSL editor for the model-driven deployment of complex applications and concepts. It supports the specification of infrastructure resources, application deployments and their tuning for performance optimization.
- FindIaCBug: easily and interactively verifies the Infrastructure as Code (IaC) and detects errors, smells, antipatterns, and bugs, while providing useful recommendations on how to fix any detected quality issue.
- Mooring: Ensures smooth deployment, orchestration, and provisioning of resources through a framework for Infrastructure as Code and modelling application deployment based on TOSCA and Ansible parameterized libraries.
- REFIT: Dynamic monitoring of systems and networks by collecting application-level and infrastructure-level metrics and events for reconfiguration and refactoring to alter deployment, quality and configuration of heterogeneous infrastructure resources used by the application.
- POET: Static and dynamic application performance optimization during design and runtime including manual profiling and tuning of application to the cloud and edge deployments based on HPC optimization.
SODALITE toolkit provides multiple benefits to application developers and infrastructure operators with tools: increasing design and runtime effectiveness of software-defined infrastructures; ensuring high-performance execution over dynamic heterogeneous execution environments; increasing simplicity of modelling applications and infrastructures; improving manageability, collaboration, and time-to-market.
To validate the performance, SODALITE tools were tested within three real-life scenarios:
- Snow water: Exploiting the operational value of information derived from public web media content with sophisticated Machine Learning techniques to assess and predict water availability coming from snow in the mountains of Lombardia.
- Vehicle IoT: Builds on ADAPTANT’s KnowGo Car data management and services platform for connected vehicles which blends automotive and personal data for the creation of data-driven vehicle services.
- In-silico clinical trials: Assessment and decision-support system for spinal operations providing efficient data access from heterogeneous compute resources and simulation process chain facilitating comprehensive data analytics.
