Bringing your AI models to EOSC Training. Register now!
AI4EOSC experts will host a training on how the project evolve innovative services to support machine learning over the European Open Science Cloud (EOSC) to better support the new challenges ahead when building distributed AI applications and improve scalability and performance for large-scale AI applications. Ignacio Heredia (CSIC), Khadijed Alibabaei (KIT), Amanda Calatrava (UPV) and Valentin Kozlov (KIT) will show how the AI4EOSC/DEEP AI platform supports typical scenarios for AI-based applications and services: re-use of existing AI modules on the marketplace, AI model development from scratch and training, and model serving. You may find more information here.This event can be attended both in-person and online and will take place on 23rd June at 09:00am CEST time. The AI4EOSC training will be a part of the upcoming EGI 2023, to be held in Poznan, Poland, 19-23 June.The registration deadline ends on 31st May 23:59 CEST, so don’t miss your chance and register now!
AI4EOSC makes its C4 models openly available in GitHub
The AI4EOSC Project has made its C4 model openly available on GitHub. A read-only version is available on Structurizr too. C4 model architecture is becoming widely popular -also when it comes to both EOSC and non EOSC-related projects- as it allows users to easily define several layers of abstraction. The C4 architecture diagrams help understanding the software system used for AI4EOSC by showing the inter-relations between the different components in the project environment.The content is licensed under a Creative Commons Attribution 4.0 International License.
The open-source platform OSCAR included in the EOSC portal
The Open-source Serverless Computing for Data-Processing Applications (OSCAR) (doi: 10.1109/CLOUD.2019.00073 ) has been included in the European Open Science Cloud (EOSC). The Grid and High Performance Computing Group (GRyCAP) at the Instituto de Instrumentación para Imagen Molecular (I3M) from the AI4EOSC consortium member Universitat Politècnica de València (UPV) developed OSCAR, a framework to efficiently support the Functions as a Services (FaaS) computing model for general-purpose file-processing applications. In AI4EOSC, the AI as a Service (AIaaS) developed will be based on the OSCAR platform, as it can be automatically deployed on multi-Clouds to create highly-parallel event-driven file-processing serverless container-based applications and execute them in response to inference requests. Official documentation: https://docs.oscar.grycap.net/ GitHub repository: https://github.com/grycap/oscar