Synopsis of cloud broker model

950 Words2 Pages

CORDS( Compatible one resource description system) is an object based description model utilizing Cloud resources. It is an open source broker system [13]. It takes the requirement and does not change anything from the Cloud provider’s side. It operates on the three primary roles: aggregator, integrator and customizer identified by Gartner[14]. Using it the broker can also switch the consumer from one Cloud to another. CORDS model acts as an important element for interpolation and federation in Cloud providers.
Based on OCCI[15] standards that defines a meta model for Cloud resources and RESTful protocol with flexible API and a strong interoperability feature, CORDS is suitable to serve IaaS, PaaS and SaaS. CORDS model is designed to be consistent and complaint with OCCI interface.In IaaS CORDS model, nodes defines a unit in terms of its infrastructure and services image. In accordance with OCCI IaaS extensions, the infrastructure unit provides Network, Storage and compute. The image unit provides the operating system. In CORDS, PaaS demonstrates an application and its environment in which it is hosted.
ACCORDS (advance capability of CORDS) are the execution platform. ACCORDS offer services of resource provisioning from different IaaS and PaaS providers through four steps:
1. Managing user’s requirement.
2. Provisioning and Validation on plans.
3. Execution of the validated plan.
4. Providing the Cloud service.
Rather than brokering, ACCORDS offers services like CompatibleOne Security Service (COSS), CompatibleOne MONitoring Service (COMONS), CompatibleOne Placement Service (COPS), CompatibleOne Energy Efficiency Service (COEES), CompatibleOneOrdering, Billing and Accounting Service (COOBAS) , CompatibleOne Network Service (CONE...

... middle of paper ...

..., KIF, SPARQL and RDQL and can be extended for providing other logical expressions. Reasoning support is limited.

The architecture of the model operates as follow. The Cloud service provider (CSP) provides product to the system. The entire product information is clustered into service repository. The CSP has to provide the domain ontology as well. The crucial part of the system is match making engine which matches the consumer service request with the services in the repository and results the best match. Input output matching is performed first followed by constraint matching. It is checked that best match obtained is Invert-Subsume.

4. Conclusion and Future Work
In this paper, we have discussed four models to serve the Cloud broker system.
Future work will be proposing a new broker system taking the best logical ideas from the other systems and implementing it.

Open Document