Newton & Infiniflow - Develop Local, Deploy Global – Next gen distributed OSGI & SCA App Server for the Enterprise

Mike Francis (), Richard Nicholson (Paremus)

OSGi DevCon · Poster

This talk has been accepted but has not yet been assigned a time slot.

Richard Nicholson

OSGi adoption in the Enterprise infrastructure software world is flourishing. Most of the commercial and open source application servers have moved, or committed to move towards the OSGi component model for componentizing and managing their own platform and supporting OSGi-based applications. Many middleware offerings are also moving this way. The adoption of OSGi by the Spring Framework provides these benefits to application developers. At the same time there is growing activity and interest around SaaS and Cloud Computing. Newton (www.codecauldron.org) and Infiniflow (www.paremus.com) provide a solution at the intersection of these areas offering an Enterprise-class distributed OSGi runtime to support OSGi-based applications across anonymous physical and virtual servers, breaking down the application silos and maximizing server utilization so solving one of the key issues and cost of datacenters today. Newton is the open source foundation of the commercial Infiniflow distributed OSGi runtime that we call a Service Fabric. They are themselves made up from OSGi bundles, and provide a highly resilient, scalable, agile platform for running OSGi-based applications and middleware. They offer the realization of Develop Local, Deploy Global. Application logic can be built and tested on a single machine and once working the code can be easily deployed in a distributed environment simply by changing the OSGi bundle bindings from local to remote with no need for any changes to the code. By automating the distribution, deployment, resilience and scalability of applications using a simple but powerful SCA-based Model Driven approach to constructing applications Developers are insulated from the complexities of distributed application design and are able to realize the full potential of component re-use and ease of application evolution allowing them to concentrate on the more interesting challenges rather than the mundane reproduction of similar functionality time and time again. At the same time Operations also benefit from these automated capabilities making it easy and efficient to install, start, pause, stop, evolve and uninstall the applications in production.

Richard, as CEO and Founder, has actively driven the strategic business and technical direction of Paremus since its formation in 2001. Richard maintains keen interest in a number of research areas including Recovery Oriented techniques, Complex Adaptive System design and Self Organized-Criticality. He is specifically interested in the application of such concepts to next generation distributed system design. Prior to founding Paremus, Richard headed the European System Engineering function for Salomon Smith Barney/Citigroup. Richard graduated from Manchester University with Honors in Physics and went on to gain an Astrophysics doctorate from the Royal Greenwich Observatory. Richard's blog can be found at: http://adaptevolve.blogspot.com/.

Gold sponsors

BEA logo

IBM logo

Wind River logo

Replay Solutions logo

JBoss logo

SOPERA logo

Cloudsmith logo

BIRT Exchange logo

Skyway Software logo

Oracle

BlackBerry logo

AMD logo

Silver sponsors

Sybase

Google

Genuitec

Instantiations

Teamprise

Telelogic

Innoopract

Business Objects

LynuxWorks logo

Hardware sponsor

AMD logo

Media sponsors

Extension Media

Methods and Tools

ACM

Addison-Wesley logo

SD Times logo

Open Systems Publishing

Software Test & Performance logo

SDForum

Be a Sponsor