IT Briefcase Exclusive Interview: Achieving a Continuum of Investment with Red HatNovember 1, 2013 No Comments
In the interview below, Pierre Fricke from Red Hat discusses challenges that arise as new technologies and tools evolve within the enterprise, and outlines best practices for organizations looking to integrate these mechanisms into their existing business processes.
- Q. In your opinion, how important is integration to open source today?
A. Integration is very important – not just to open source, but IT and business in general. Technology innovation is booming in arenas such as social, mobile, cloud, and big data, and much of this innovation is happening in open source. These innovations are developed in communities, experimented with, and deployed in new architectures to solve IT and business problems in new ways. However, if they cannot be integrated into IT and the business, their impact is limited. This means that these technologies must not only work themselves and on their own data, but must integrate and leverage previous IT investments in applications and data to be effective. They also must be integrated into business processes (new or existing) to benefit the business by helping it to run better, faster and smarter.
- Q. How do big data and cloud computing create new integration challenges for enterprises today?
A. Big data brings a much wider range of data formats, larger quantities, and greater variability in speed of arrival of the data than before. This variety of format, quantity and speed present new challenges that the old-style database connectors really do not solve. For example, to make the best decision about a loan application or to respond to a business event, applications/people need all of the most current, relevant data presented in a useful form. This requires high-speed, distributed messaging and integration, couple with data virtualization, to give the application/people the exact information they need to make an optimal decision. The decision itself may well be codified in a business rules management system (BRMS) which enables the business team to make updates and view/audit the decision making process.
Likewise, the cloud presents new architectures that span firewalls, communities and networks, and add variation to the applications and data that need to be integrated with each other and into the business. These new architectures are built from IaaS and PaaS infrastructure that automate and abstract lower levels of the IT stack, away from applications and business processes. Integration infrastructure must not only run on native containers, but also private, public and hybrid cloud-based containers. Integration infrastructure can benefit from this cloud infrastructure as well – e.g., scalable underlying infrastructure to run widely-variable message and data flows through the integration fabric. The integration infrastructure must be able to get messages in a variety of formats (e.g., AMPQ, MQTT, HTTP, SOAP) through the firewall. The integration development environment must be integrated into the PaaS developer tool chain for maximum development and operational efficiency. Integration infrastructure that can do these can be considered Integration PaaS (iPaaS).
- Q. Can you explain a little about the concepts of “big data at rest” and “big data in motion”?
A. Today, there is a proliferation of smart devices, sensors, and intelligent mobile assets that collect and/or generate data on a continual basis. This data is streamed into either key business applications or business processes to enable smarter, faster execution of business. These applications, rules and decision management systems, and business processes use this data in motion to make real-time decisions based on up-to-the-second live information. This is “big data in motion”.
“Big data at rest” involves traditional relational, flat file, spreadsheet, and other data that is coupled with new forms of data that is generated by the smart network described above. The new forms of data that arrive from the field may be processed in motion as described and then stored in repositories such as Hadoop or MongoDB for further analysis after the fact. This data becomes “big data at rest”.
- Q. How has this affected the data management solutions Red Hat is putting out today, and what role does social big data play in this picture?
A. This is best described in an announcement this week around big data…
We are adding big data integration to our JBoss Data Virtualization offering with Apache Hive connector to Hadoop, MongoDB connectivity and JBoss Data Grid integration. ( see http://www.redhat.com/about/news/archive/2013/10/red-hat-turns-big-data-into-actionable-information-with-jboss-data-virtualization ).
Our JBoss A-MQ messaging platform, based on ActiveMQ, enables integration of these smart devices, sensors and mobile assets into applications, rules management systems and business process management to process, respond to and even anticipate events.
JBoss BRMS provides intelligent decision making by interpreting these big data in motion streams (and/or big data at rest) against a set of business rules.
JBoss Data Virtualization presents a consumable view of needed data (traditional, big data, etc…) to these applications, and BRMS/BPMS systems.
Social data is a form of both big data in motion as it is happening and big data at rest as it is stored. Social data provides valuable insight into trends, sentiment, opportunities and problems associated with launches, products, services, events in real time. Integrating and analyzing this social data can significant improve response and even enable preventative or opportunistic business actions to be deployed. For example, if social like trends favor a particular color or style of a product in certain locations as it is being launched with initial purchases happening, the business can signal its supply network to shift production and delivery to the point of opportunity much more rapidly than after the fact analysis a week later.
- Q. As new technologies and tools appear on the scene daily, how is Red Hat working to help organizations integrate these mechanisms into their existing business processes, and achieve a “continuum of investment”?
A. JBoss Fuse is at the center of our integration strategy and the Apache Camel project is a major element of JBoss Fuse. Camel has a large community of contributors delivering a wide range of connectors and capabilities to integrate different messaging, data and applications. This robust open source community is the best way to keep up with all of these changes vs. a proprietary, internally developed integration platform. JBoss Fuse has a great ability to enable developers to rapidly develop and deploy integration between existing and new IT assets to solve new business problems. This rapid integration adds value to the previous investments by bringing them into new business processes achieving a continuum of investment from the past to the future.
Beyond that, JBoss A-MQ messaging delivers the foundation for Red Hat based integration within the enterprise, out to the extended enterprise and into the cloud. JBoss Data Virtualization brings a wide range of data sources together to enable a holistic or required view of a customer, patient, part, product, entity of interest, etc. The list of data sources grows with the evolution of big data.
Our forthcoming JBoss Fuse Service Works builds on JBoss Fuse enabling rapid development and reuse of integrated composite business services to support easier integrationof these new and traditional applications into business processes (see http://www.redhat.com/about/news/archive/2013/10/red-hat-debuts-its-next-generation-integration-platform-with-jboss-fuse-service-works ).
Finally, our forthcoming JBoss BPM Suite will provide the tools for business and IT users to collaborate to build new and automation existing business processes leveraging these new services in JBoss Fuse Service Works, data sources from JBoss Data Virtualization, messages from across the enterprise via JBoss A-MQ and new applications developed on JBoss Enterprise Application Server.
Click here to view mobile, cloud, and open source PAAS solutions that Red Hat is currently offering to help organizations overcome integration challenges.
As Director of Product Marketing for Red Hat’s JBoss integration and BPM products, Pierre Fricke is responsible for driving the strategy and enterprise messaging for these Red Hat products. Pierre led the product strategy and expansion into the integration and SOA market with the JBoss Enterprise SOA Platform which included JBoss ESB, registry, workflow and a rules engine. In 2009, Pierre led the launch of JBoss Enterprise BRMS to lay the foundation for a complete open source BPM strategy. Pierre was a key leader working on the FuseSource and Polymita acquisitions which are now an integral part of Red Hat’s integration and BPM strategy and product lines. Today, these products are the unit volume market leader or emerging strong challengers to long time incumbents. Pierre’s focus is to expand Red Hat’s market presence in integration, service-oriented architecture and business process automation, bringing the value of open source and community innovation to customers in these markets.
APPLICATION INTEGRATION, Fresh Ink, OPEN SOURCE