![12/17 25 – 30 June, 2017, Honolulu, Hawaii, USA
Multi-Cloud Interoperability
Modelling SpaceProvider Space Programming Space Semantic Space
TOSCAModelling
Space
CloudML OCCIware
Programming
Space GCP
SDK
AWS
SDK
Provider
Space
Public PrivateHybrid
Cloud
provider
AWS
API
OCCI
API
Public
GCP
API
Broker
Semantic
Space fcloudsPSIF [2][1] Cloud
architect
Multi-cloud
Libraries
Cloud
architect
Cloud
developer](https://image.slidesharecdn.com/challitacloud2017-170620125714/85/Towards-Formal-based-Semantic-Interoperability-in-Multi-Clouds-CLOUD-2017-12-320.jpg)
![13/17 25 – 30 June, 2017, Honolulu, Hawaii, USA
[1] K. Yongsiriwit, M. Sellami, and W. Gaaloul, “A Semantic Framework Supporting Cloud Resource Descriptions
Interoperability,” in 2016 IEEE 9th International Conference on Cloud Computing (CLOUD). IEEE, 2016, pp. 585–592.
[2] N. Loutas, E. Kamateri, and K. Tarabanis, “A Semantic Interoperability Framework for Cloud Platform as a
Service,” in 2011 IEEE Third International Conference on Cloud Computing Technology and Science (CloudCom). IEEE,
2011, pp. 280–287.
The fclouds Framework
fclouds = catalogue of formal models + transformation rules](https://image.slidesharecdn.com/challitacloud2017-170620125714/85/Towards-Formal-based-Semantic-Interoperability-in-Multi-Clouds-CLOUD-2017-13-320.jpg)
Towards Formal-based Semantic Interoperability in Multi-Clouds (CLOUD'2017)
- 1. Towards Formal-based Semantic Interoperability in Multi-Clouds Stéphanie Challita | Fawaz Paraiso | Philippe Merle Inria Lille – Nord Europe & University of Lille, France 10th IEEE International Conference on Cloud Computing (CLOUD 2017)
- 2. 2/17 25 – 30 June, 2017, Honolulu, Hawaii, USA ContextContext But the cloud solutions are heterogeneous… Misunderstandings, errors…
- 3. 3/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling Space Cloud architect Programming Space Provider Space Cloud provider Semantic Space Cloud developer Cloud architect Our classification of multi-cloud solutions
- 4. 4/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Deployment Model Management Interface Service Model Public SOAP IaaS, PaaS, SaaS Public REST IaaS, PaaS, SaaS Hybrid REST IaaS, PaaS Private REST IaaS Modelling SpaceProvider Space Programming Space Semantic Space
- 5. 5/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Developer Natural language is ambiguous! Modelling SpaceProvider Space Programming Space Semantic Space
- 6. 6/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space Provider Space Public PrivateHybrid Cloud provider AWS API OCCI API Public GCP API Broker The broker offers a new API to handle the heterogeneity of different APIs But it only masks the problem in this space and does not semantically resolve it!
- 7. 7/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space Each cloud offers one or several language-specific SDKs to hide technical details of APIs SDKs are heterogeneous! Programming Space Cloud developerGCP SDK AWS SDK Provider Space Public PrivateHybrid Cloud provider AWS API OCCI API Public GCP API Broker Multi-cloud Libraries
- 8. 8/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space The developer needs to focus on cloud general properties before a commitment to implementation choices Libraries are tightly coupled to programming languages! Programming Space GCP SDK AWS SDK Provider Space Public PrivateHybrid Cloud provider AWS API OCCI API Public GCP API Broker Multi-cloud Libraries Cloud developer
- 9. 9/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space Cloud architects need to design their applications for multi-clouds Runtime Multi-cloud Metamodel Configuration Model conforms to represented by Cloud architect defines
- 10. 10/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space The lack of formalisation hinders the understanding of the models TOSCAModelling Space CloudML Cloud architect OCCIware Programming Space GCP SDK AWS SDK Provider Space Public PrivateHybrid Cloud provider AWS API OCCI API Public GCP API Broker + OCL Invariants/ Natural Language Metamodels are heterogeneous & informal semantics is ambiguous! Multi-Cloud Metamodel Configuration Model conforms to represented by Runtime Multi-cloud Libraries Cloud developer
- 11. 11/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space Interoperability solutions must adhere to a precise set of principles Mathematical specification Formal Validation Reasoning One interpretation Accuracy Earlier error detection, cheaper correction
- 12. 12/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Multi-Cloud Interoperability Modelling SpaceProvider Space Programming Space Semantic Space TOSCAModelling Space CloudML OCCIware Programming Space GCP SDK AWS SDK Provider Space Public PrivateHybrid Cloud provider AWS API OCCI API Public GCP API Broker Semantic Space fcloudsPSIF [2][1] Cloud architect Multi-cloud Libraries Cloud architect Cloud developer
- 13. 13/17 25 – 30 June, 2017, Honolulu, Hawaii, USA [1] K. Yongsiriwit, M. Sellami, and W. Gaaloul, “A Semantic Framework Supporting Cloud Resource Descriptions Interoperability,” in 2016 IEEE 9th International Conference on Cloud Computing (CLOUD). IEEE, 2016, pp. 585–592. [2] N. Loutas, E. Kamateri, and K. Tarabanis, “A Semantic Interoperability Framework for Cloud Platform as a Service,” in 2011 IEEE Third International Conference on Cloud Computing Technology and Science (CloudCom). IEEE, 2011, pp. 280–287. The fclouds Framework fclouds = catalogue of formal models + transformation rules
- 14. The fclouds Framework OCCI API fOCCI: OCCI Formal Model Modelling fGCP: GCP Formal Model GCP API Semantic Interoperability Implements Implements 1 2 3 1’ Transformation rules Bridge Modelling 14/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Usage Scenario
- 15. 15/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Alloy Analyzer Main steps for formalizing cloud APIs Definition and Validation of cloud structure and constraints Definition and Validation of cloud API operations Textual description of cloud operations Textual description of cloud concepts Modelling Web Crawler Web Crawler VM Storage Network Container Server Create Retrieve Update Delete The fclouds Framework 1
- 16. 16/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Alloy Analyzer Main steps for formalizing cloud APIs Definition and Validation of cloud properties 2 Reasoning Definition and Validation of cloud structure and constraints Definition and Validation of cloud API operations Textual description of cloud operations Textual description of cloud concepts Modelling Web Crawler Web Crawler VM Storage Network Container Server Create Retrieve Update Delete The fclouds Framework 1
- 17. 17/17 25 – 30 June, 2017, Honolulu, Hawaii, USA Future Work Implement formal cloud models fOCCI, fGCP, fTOSCA, etc. Define & verify cloud properties Define transformation functions between fclouds models Complete the fclouds framework