SODALITE

SOftware Defined AppLication Infrastructures managemenT and Engineering

SODALITE (SOftware Defined AppLication Infrastructures managemenT and Engineering) is aimed at tackling the complexity of deploying and operating modern applications onto heterogeneous HPC and cloud-based software-defined infrastructures, under arbitrary operational conditions and requirements. This challenge is addressed through (a) the abstraction of the application and software-defined infrastructures, (b) the automation of maintaining a dynamic balance between the application and the infrastructure, and (c) the design and runtime analysis of software-defined infrastructures. The novelty of the SODALITE’s approach lies in developing and using an application, infrastructure, and performance abstraction library to enable dynamically optimized multi-target application execution via an optimal mixture of Cloud, HPC, GPUs, FPGAs, etc., based on the requirements of the application itself and the dynamic state of the infrastructure. Further on, dynamic optimization will be based on continuous application and infrastructure monitoring, to enable radical on-the-fly reconfigurations of the deployed applications.

SODALITE envisions supporting the Digital Transformation of European Industry through (1) increasing design and runtime effectiveness of software-defined infrastructures, to ensure high-performance execution over dynamic heterogeneous execution environments; (2) increasing simplicity of modelling applications and infrastructures, to improve manageability, collaboration, and time to market.

Within the project, MKLab as a department of the Information Technologies Institute (ITI) of CERTH will lead WP3, dealing with the semantic abstractions and the relevant design and modelling of applications and cloud infrastructures, along with their performance characteristics and deployments. Last but not least, MKLab will participate and contribute, to various extents, to all WPs (WP1-WP7).

PUBLICATIONS

  1. Indika Kumara, Alfio Lazzaro, Nina Mujkanovic, Zoe Vasileiou & Damian A. Tamburri, "Quality Assurance and Design-Time Optimization" In Elisabetta Di Nitto, Jesús Gorroñogoitia Cruz, Indika Kumara, Dragan Radolović, Kamil Tokmakov, Zoe Vasileiou (eds), Deployment and Operation of Complex Software in Heterogeneous Execution Environments , Springer, DOI:https://doi.org/10.1007/978-3-031-04961-3, 53-66, July 2022.
  2. Jesús Gorroñogoitia, Dragan Radolović, Zoe Vasileiou, Georgios Meditskos, Anastasios Karakostas, Stefanos Vrochidis & Michail Bachras (2022), "The SODALITE Model-Driven Approach" In Elisabetta Di Nitto, Jesús Gorroñogoitia Cruz, Indika Kumara, Dragan Radolović, Kamil Tokmakov, Zoe Vasileiou (eds), Deployment and Operation of Complex Software in Heterogeneous Execution Environments , Springer, DOI:https://doi.org/10.1007/978-3-031-04961-3, 23-52, July 2022.
  3. G. Meditskos, Z. Vasileiou, A. Karakostas, S. Vrochidis, I. Kompatsiaris, “A Pattern-based Semantic Lifting of Cloud and HPC Applications using OWL 2 Meta-modelling”, International Conference on High Performance Computing & Simulation (HPCS 2020), Barcelona Spain, 10-14 December 2020.
  4. I. Kumara, Z. Vasileiou, G. Meditskos, D. Tamburri, W.J. Van Den Heuvel, A. Karakostas, S. Vrochidis and I. Kompatsiaris, "Towards Semantic Detection of Smells in Cloud Infrastructure Code", 10th International Conference on Web Intelligence, Mining and Semantics (WIMS2020). OPEN ACCESS