Periodic Reporting for period 1 - PHYSICS (Optimized hybrid space-time service continuum in FAAS)
Reporting period: 2021-01-01 to 2022-06-30
The current Cloud computing landscape is characterized by an extreme diversity of offerings and services, incorporating multiple solutions from centralized Cloud providers (such as typical VM offerings, dedicated nodes, hardware enhanced resources such as GPUs and FPGAs etc), edge and fog environments, HPC facilities, mobile computing applications etc. implementing the Everything as a Service approach.
On the other hand, applications are typically consisted of a multitude of components, others in need of locality and others in need of significant computational resources to fulfil their scope and objectives, with varying abilities to exploit the underlying services, varying requirements for operation and control as well as technologies and programming structures/languages on which they rely. PHYSICS empowers European CSPs (Cloud Solution Providers) exploit the most modern, scalable and cost-effective cloud model (FaaS), operated across multiple service and hardware types, provider locations, edge, and multi-cloud resources. To this end, it applies a unified continuum approach, including functional and operational management across sites and service stacks, performance through the relativity of space (location of execution) and time (of execution), enhanced by semantics of application components and services. PHYSICS aims at delivering a complete vertical solution that will enable:
a) CSPs to offer advanced cloud application design environments to their main end customers (Application Developers) to abstractively create workflows of their applications, exploiting generalized Cloud design patterns for functionality enhancement with existing application components, easily designed and reused through intuitive visual flow programming tools (CSP Cloud Design Environment);
b) Platform-level functionalities to be easily incorporated by providers in order to translate the created application workflows to deployable functional sequences, based on the Function as a Service model, optimizing their placement across the Cloud computing domain and exploiting the computational space-time continuum as well as advanced semantics for the definition of a global service graph (CSP Optimized Platform Level FaaS Services Toolkit);
c) Provider-local resource management mechanisms that will enable providers to offer competitive and optimized services with extended interfaces offering local fine grained control of elasticity rules and policies, while applying a holistic set of provider-local strategies based on a wide set of controlling techniques and tackling key aspects of multitenancy (CSP Backend Optimization Toolkit). PHYSICS will contribute to open source tools and initiatives/policies (Gaia-X, Green Deal, EOSC, Eur. Strategy for Data), while validating the outcomes in 3 real-world applications (eHealth, Agriculture and Manufacturing), making a business, societal and environmental impact on EU citizen life.
On the other hand, applications are typically consisted of a multitude of components, others in need of locality and others in need of significant computational resources to fulfil their scope and objectives, with varying abilities to exploit the underlying services, varying requirements for operation and control as well as technologies and programming structures/languages on which they rely. PHYSICS empowers European CSPs (Cloud Solution Providers) exploit the most modern, scalable and cost-effective cloud model (FaaS), operated across multiple service and hardware types, provider locations, edge, and multi-cloud resources. To this end, it applies a unified continuum approach, including functional and operational management across sites and service stacks, performance through the relativity of space (location of execution) and time (of execution), enhanced by semantics of application components and services. PHYSICS aims at delivering a complete vertical solution that will enable:
a) CSPs to offer advanced cloud application design environments to their main end customers (Application Developers) to abstractively create workflows of their applications, exploiting generalized Cloud design patterns for functionality enhancement with existing application components, easily designed and reused through intuitive visual flow programming tools (CSP Cloud Design Environment);
b) Platform-level functionalities to be easily incorporated by providers in order to translate the created application workflows to deployable functional sequences, based on the Function as a Service model, optimizing their placement across the Cloud computing domain and exploiting the computational space-time continuum as well as advanced semantics for the definition of a global service graph (CSP Optimized Platform Level FaaS Services Toolkit);
c) Provider-local resource management mechanisms that will enable providers to offer competitive and optimized services with extended interfaces offering local fine grained control of elasticity rules and policies, while applying a holistic set of provider-local strategies based on a wide set of controlling techniques and tackling key aspects of multitenancy (CSP Backend Optimization Toolkit). PHYSICS will contribute to open source tools and initiatives/policies (Gaia-X, Green Deal, EOSC, Eur. Strategy for Data), while validating the outcomes in 3 real-world applications (eHealth, Agriculture and Manufacturing), making a business, societal and environmental impact on EU citizen life.
First versions of the PHYSICS Application Core ontology, Application Extension ontology, Service Semantics ontology, Application graph, and Workflow app definition.
First Version of the Design Environment.
First release of the architecture.
Design and development of the Workflow Operator CRD component.
Integration flow between technical WP components and between WPs defined, through APIs, in a kubernetes native way.
Three pilot use cases described in a detailed Use-Case scenarios analysis.
Each pilot set up their local experimentation environment and infrastructure and adapted the first prototype applications.
Cloud Infrastructure for PHYSICS and CI/CD agile processes and pipelines selected, defined and set up.
Integration of all the PHYSICS components into a solution framework (PHYSICS MVP) that provides all required services (with deployment on the target infrastructure).
Reusable Artefacts Marketplace Platform (RAMP) v1.0 deployed and released.
First Version of the Design Environment.
First release of the architecture.
Design and development of the Workflow Operator CRD component.
Integration flow between technical WP components and between WPs defined, through APIs, in a kubernetes native way.
Three pilot use cases described in a detailed Use-Case scenarios analysis.
Each pilot set up their local experimentation environment and infrastructure and adapted the first prototype applications.
Cloud Infrastructure for PHYSICS and CI/CD agile processes and pipelines selected, defined and set up.
Integration of all the PHYSICS components into a solution framework (PHYSICS MVP) that provides all required services (with deployment on the target infrastructure).
Reusable Artefacts Marketplace Platform (RAMP) v1.0 deployed and released.
Looking at the opportunities and solutions, potential viable options regarding exploitation plans include leveraging exploitable results both as a whole thanks to the 3 bundles approach working in synergy with the RAMP, or individually through 3 bundle-specific exploitation paths with the RAMP also working independently. In this sense, the Cloud Design Environment is designed for CSPs that lack advanced workflow definition, the Optimized Platform Level FaaS Services Toolkit targets CSPs who wish to extend their services from IaaS to FaaS, and the Backend Optimization Toolkit focuses on the CSPs that wish to enrich their currently sophisticated offering. On the other hand, the RAMP aims at bringing contributors
and buyers from the cloud computing environment around one central artefacts marketplace.
and buyers from the cloud computing environment around one central artefacts marketplace.