Paremus Service Fabric

The Private Cloud for your composite applications

The Paremus Service Fabric offers a distributed, dynamic, resilient and scalable Private Cloud runtime for your composite applications.

OSGi™ from the ground-up, the Paremus Service Fabric is itself formed from a collection of Nimble-enabled compute nodes.

Standards Based

The Paremus Service Fabric supports OSGi Alliance standards including Declarative Services and the Blueprint Service.

The Service Fabric also ships with a high performance low latency implementation of the Remote Service Admin (RSA) standard which uses OMG/DDS based discovery.

DDS is also used to provide a low latency command and control backplane providing a Private Cloud for most demanding of environments.

Driving Down Application TCO

The Paremus Service Fabric is model-driven. This enables it to automatically manage the provisioning and on-going maintenance of the simplest to the most sophisticated composite applications, maintaining all of the components and the runtime infrastructure upon which they depend.

Each application is completely described by its corresponding System model. No externalconfiguration management (CMDB) or deployment solutions are required.

To re-configure a running application, simply make the appropriate configuration changes to the corresponding System model and these are then automatically propagated to all runtime components.

To patch a running application, simply create a new version of the System model with the updated components and then activate this. The Service Fabric responds by upgrading the affected software components within the running application. This upgrade is achieved within seconds.

To rollback; simple re-activate the previous version of the System model and the running application will be automatically rolled-back within seconds.

Finally, when presented with a software component, each Service Fabric Fibre, responds by determining all local runtime dependencies and dynamically installing all of the required artifacts to fulfill the dependencies.

For further details on supported artifact types see the Nimble Resolver.

Robustness Through Recovery

The Paremus Service Fabric is a truly robust platform that treats failure, recovery and environmental change as normal occurrences rather than unexpected exceptions.

The Service Fabric monitors each running composite application, using the corresponding System model as the point of reference. When anomalies are detected, the Service Fabric takes the appropriate corrective action, e.g. automatically re-provisioning software components that were lost due to infrastructure failure.

The Service Fabric automatically adapts to changes in both the operational environment and business conditions. As long as suitable resources are available, the Service Fabric is able to re-heal its own command and control subsystems, so ensuring that at each point in time there are no-single points of failure.

Middleware Agnostic

The Paremus Service Fabric does not enforce any particular architectural form on the hosted applications.

Messaging middleware and protocols are the concerns of each System.

The messaging middleware and protocols are determined by the selection of bindings and middleware components used within each System e.g. SEDA, ESB, Tuple Space or P2P.

Low latency, batch grid, order matching and transactional messaging based systems can all be run concurrently on the same Service Fabric.

Likewise, the Service Fabric does not dictate persistence API’s, instead it allows the selection of the most appropriate NoSQL solution based upon the requirements of your business data and the Consistency, Availability and Partitionability trade-offs dictated by your business SLAs.

To help get you started we provide a number of popular open source and commercial middleware and NoSQL technologies as optional pre-packaged Runtime Patterns. We also offer consulting services to help you integrate alternatives.