Announcing: Red Hat Fuse 7 is now available

After several technical previews over the last few months, Red Hat Fuse is officially available. This is a significant release, both for Fuse itself and for integration platforms, because it represents a shift from more traditional, on-premise, centralized integration architecture to distributed, hybrid environment integration architecture.

Integration itself has historically been a bottleneck for infrastructure design and changes. The integration points were largely centralized and controlled by a central team in an attempt to manage dependencies and standardize data management between applications. However, that centralization also made change difficult, and it was governed more by procedure and bureaucracy than business innovation. As with traditional infrastructure architecture more generally, integration has not historically been an agile or adaptive architecture.

Red Hat Fuse (and related community projects) is the beginning of a departure from traditional, rigid integration platforms to more agile, distributed integration design. Fuse introduces three major features in the latest release:

  • Fuse Online, fully hosted Fuse applications and integrations. Fuse Online provides immediate access to the functionality of Fuse, without having to install and configure it on-premise. Developers can begin testing and customizing integrations immediately. Connectors can be uploaded to the online development area to allow even more integrations.
  • Fuse container images for Red Hat OpenShift. Fuse runs natively on OpenShift, allowing local, containerized integration points to be created in development teams and to be designed, tested, and updated within DevOps workflows as part of the overall application development cycle.
  • A drag-and-drop UI for integration pattern design. While integration development is typically done within IT teams, integration design relies on business knowledge. Business managers and analysts need to be able to collaborate effectively with their development teams. The new Fuse Ignite UI (based on the Syndesis.io project) is a lowcode way to develop integration — business users can use design elements to create integration architectures and to work with their development teams, within the same tool set.

These three features allow more agile integration development. Fuse installations can span online, on-premise, or container based environments without reducing functionality. This allows an integration platform that crosses environments, and be as lightweight and decentralized as an individual development team or an enterprise-wide platform. The lowcode UI allows business users to be brought directly into the application development cycle, enabling business logic to be incorporated into the integration application design from the beginning.

Additionally, Fuse 7 contains these new features:

  • Support for Spring Boot deployment for Fuse applications
  • 50 new application connectors (with a total of over 200 included connectors)
  • A new monitoring subsystem
  • Updated component versions, including new versions of Red Hat JBoss Enterprise Application Platform and Apache Camel
  • A new name (Red Hat Fuse, rather than Red Hat JBoss Fuse)

 

Additional Resources

#RHSummit: A Random Sampling of Awesome Sessions and Events Throughout the Week

There are around 500 sessions crammed into a speedy three day schedule — so it is impossible to catch everything. (That’s one reason that I’m promoting things like theCube streaming channel and recorded sessions on our Youtube channel — it’s a way to catch all the things you can miss, even if you attend something every hour.)

If you haven’t already mapped out everything to see and do, the trailmaps are a great place to start to get the cream of every topic area.

I have created my own, unscientific list of the app dev and middleware-related sessions that caught my eye in the session catalog.

Stuff to Do

There are after-parties most nights, some on site at the Moscone Center and some at the conference hotels. Keep an eye on the signs in the lobbies — there are lists there. For those passionate about app development, middleware, and application architecture:

  • There will be a press panel including Mike Piech (VP of middleware) and Harry Mower (Red Hat Developers) in the Intercontinental Hotel Ballroom A. Space is limited, so it will also be broadcast live on theCube at 11am.
  • There is a rockstar cocktail hour on Wednesday evening, starting at 5:30 in Moscone West.
  • Also on Wednesday, Mike Piech and Mark LIttle will do an interview with theCube. Along with streaming live online, you can see it in person in the Moscone West lobby.
  • The Red Hat Summit wrap party is Thursday night at the Armory, starting at 7pm.

Integration

Trailmap: Integration

Distributed API management in a hybrid cloud environment
Tuesday, 10:30am, Moscone West 2003
Why it’s cool:
This is a real customer story on how they used API management with 3scale to manage thousands of services across a hybrid environment.

Practitioner’s guide to API strategy
Wednesday, 3:30pm, Moscone South 207
Why it’s cool:
Anything with “strategy” in it catches my eye. This session goes over why and how an API initiative should be structured to be successful.

Introducing AMQ streams – data streaming with Apache Kafka
Thursday, 11:15am, Moscone West 2014
Why it’s cool:
Microservices — or any kind of distributed computing system — comes down to a question of managing data. This looks at some new technologies in AMQ so that the messaging platform can span a variety of data architectures, from IoT to enterprise integration to (also) data streaming.

Decompose a monolith with microservices
Thursday, 3:00pm, Moscone West Level 2, Discovery Zone
Why it’s cool:
 Another session hitting the same point — distributed architectures are complex. You need a clear understanding of interdependencies, integration points, and data (among many other things), and this session breaks down what you need to know and best practices for addressing it.

Future Technologies

There are a lot of separate, and separately interesting, technologies on the horizon. The ones that seem to stick out at this Summit revolve around serverless computing or Istio Service Mesh.

Containers, microservices, serverless: On being serverless or serverful
Tuesday, 10:30am, Moscone South 207
Why it’s cool:
Burr Sutter presenting plus serverless and microservices in the title.

Istio: Solving the challenges of hybrid cloud
Tuesday, 3:30pm, Moscone South 208
Why it’s cool:
 This goes over how Istio can be used in an infrastructure that spans OpenShift containers, Kubernetes, and virtual machines. Managing data across environments effectively is a major challenge as applications and services need to be able to scale.

Low-risk mono to microservices: Istio, Teiid, and Spring Boot
Tuesday, 4:30pm, Moscone South 207
Why it’s cool:
This looks at how to break a monolith — fully recognizing that there are no clear-cut boundaries in a monolith and the interdependencies get messy.

An eventful tour from enterprise integration to serverless computing
Wednesday, 10:30am, Moscone South 207
Why it’s cool:
This looks at the different architectural designs and choices for event-driven computing, microservices, messaging, and data management. There isn’t a single perfect solution that works for everyone — each infrastructure has its own priorities and needs, and those have to be reflected in the architecture.

Internet of Things

Trailmap: IoT

Making IoT real across industries
Tuesday, 11:45am, Moscone West 2007
Why it’s cool:
Tell me a story. IoT is essentially a highly complex integration story, integrating not only applications but physical devices. Three different industries — technology, petroleum, and transportation — highlight different aspects of IoT as it was done in real life.

Internet of Things: Open, integrated, managed, and secure
Thursday, 3:00pm, Moscone West 2016
Why it’s cool:
How do you take a cool idea (or a business necessity) and make it happen in real life? This section includes common reference architectures for industrial IoT deployments.

Cloud-native and App Dev

Trailmap: Cloud-native apps

Containerizing applications — existing and new
Wedneesday, 1:00pm, Moscone South 155
Why it’s cool:
Anything practical is immediately appealing. Most organizations aren’t dealing with a greenfield of applications, and this looks at how to move both cloud-native and legacy applications into a container.

Using machine learning, Red Hat BPM, and reactive microservices
Thursday, 11:15am, Moscone West 2004
Why it’s cool:
Business process automation, decision management, event processing — these tend to be treated as commodity actions. The things you have to do to get an application to be more responsive with less intervention. I like the approach of adding machine learning to process management, giving more intelligence to the overall architecture.

Java Awesomeness

Eclipse Microprofile and Wildfly Swarm
Tuesday, 11:45am, Moscone West 2011
Why it’s cool:
There isn’t a ton of Java on this lit (I don’t really know why), but this is definitely a don’t-miss session for Java developers. Wildfly Swarm is a way to create cloud-native, container-native Java applications. So … all your Java expertise, in a tiny container.

EE4J, MicroProfile, and the future of enterprise Java
Wednesday, 3:30pm, Moscone South 215
Why it’s cool:
 There are probably a dozen think-pieces a year on the imminent death of Java — yet it continues to evolve across new architectures and to take on new technologies. This session takes a more optimistic (realistic?) view of the future of Java.

Microservices data patterns: CQRS and event sourcing
Thursday, 11:15am, Moscone South 208
Why it’s cool:
Microservices (as Christian Posta is fond of saying) represent a data challenge. The more distributed the data is, than the more structured and clear the data architecture needs to be.

Crossing the chasm between traditional and agile planning
Tuesday, 1:45pm, Moscone West 2103
Why it’s cool:
Teams are people. Technology has to be developed and executed and maintained by people. Making any kind of shift, whether changing the planning structure or the infrastructure architecture or something else, requires an understanding of how to manage and inspire teams.

 

#RHSummit: Plug in (whether you’re here or not)

Red Hat Summit — the unofficial start of summer technology discussions and the official conference for all things open source — begins tomorrow. For attendees, there are a handful of links to keep handy so that you can hit the sessions, booths, demos, and after-hours events that make Summit so awesome.

Even if you can’t attend this year (or since, realistically, no human being can attend everything going on at Summit), here is a round-up of social media channels and people to follow so you can dip your toe into the Summit experience.

**NEW** Interesting Personal Accounts

A lot of the people presenting at Summit or working in the DevZone and Partner booths have a social media presence all their own. It’s definitely worth tracking what they’re doing at Summit, and after. A handful:

Live-streaming Summit

All of the Red Hat Summit general sessions will be live-streamed on theCube, along with interviews and round tables throughout the day. Previous years’ Summits are also available in theCube archives or on our Red Hat Summit YouTube channel.

Live streams of note:

  • Morning general sessions, Tuesday and Wednesday, 8:30am
  • Press conference, with live Q&A, Tuesday, 11am
  • Afternoon general sessions, Tuesday and Wednesday, 1:45pm
  • The Future of Java interview with Mike Piech and Mark Little, Wednesday
  • Closing general session, Thursday, 8:30

Summit-Specific Accounts

Follow the #RHSummit hastag — use it to be part of the conversation.

Middleware-Related Social Media

Main Red Hat Sites

“Micro-rules,” event-driven apps, and Red Hat Decision Manager

As we described in an earlier blog, microservices are mini-applications which are devoted to a single, specific function. They are discrete (independent of other services in the architecture), polyglot with a common messaging or API interface, and they have well-defined parameters.

As application development and IT operations teams have started streamlining and speeding up their processes with methodologies like Agile and DevOps, they have increasingly begun treating IT applications as microservices. This breaks up potential bottlenecks, reduces dependencies on services used by other teams, and can help make IT infrastructure less rigid and more distributed.

One area where we are seeing this looser, more distributed approach to service development is with business rules.

“Micro-rules”

Business rules and processes in a traditional structure tend to be centralized, with the complete set of functionality defined for all workflows. The problem with centralization is because there is a single, centralized collection of business rules, any changes to one set of rules can affect many other sets, even those for different business functions.

Micro-rules essentially treat each functional set of rules as its own service — well-defined, highly focused, and independent of other rules.

Figure – Function rule sets as micro-rules

Continue reading ““Micro-rules,” event-driven apps, and Red Hat Decision Manager”

Announcing: Red Hat Decision Manager 7.0 Is Now Available

Red Hat has announced the release of Red Hat Decision Manager 7. Decision Manager is the evolution of Red Hat JBoss BRMS and provides a platform to develop rules-based applications and services.

As applications and services become more central to business strategies, business users will become increasingly involved in the development process. Software that aids in creating applications without directly writing code is known as low-code development. Decision Manager provides tools, including an updated UI and enhanced wizards, that help business users participate more actively in application development.

Major Use Cases

Decision service as a microservice

Decision Manager has a more modular architecture, such as a decision service, an execution server, and a management interface. Each component can be containerized and deployed as an image on Red Hat OpenShift Container Platform. Developers can create discrete services for specific needs and deploy those “micro-rules” in a microservices architecture. This approach is covered more in another blog post.

Continue reading “Announcing: Red Hat Decision Manager 7.0 Is Now Available”

What is agile integration?

If you Google the term “agile integration,” you’ll come up with about 30 million results, but they focus heavily on one area: continuous integration within agile development. That definition of agile integration is based on the build environment.

However, it is possible to have another definition for “agile integration,” one that looks at the platform architecture.

In this definition, “agile” doesn’t relate to the process or the infrastructure, but to the flexibility and adaptability–the agility–of the application architecture. Integration within this context has a more strategic role, as the architectural framework that defines the interoperability of services and with a focus on the application functionality.

Traditional vs. agile as an architectural approach

There are functional similarities between traditional integration and agile integration – like routing, connectivity, and orchestration capabilities. The difference between traditional enterprise application integration and agile integration is not in the tasks performed, but in the strategic perspective of those tasks. Put simply, integration can be viewed as a necessary but often limited part of the infrastructure (traditional) or it could be viewed as the core framework of the application architecture (agile).

Continue reading “What is agile integration?”

Summit Notes: Tuesday Morning General Session

If you missed it, the keynote speeches are available on the Summit page or on YouTube.

“You don’t need to focus on technology. You need to empower your developers.”

There are certain patterns in the middleware / application development tracks for Red Hat Summit this year, and they revolve a lot around microservices. That makes a certain kind of sense (microservices are the new hotness in app development), but it’s also reflective of a larger current in technology, a continuing push toward … something.

In his opening keynote, Red Hat EVP Paul Cormier noted that one of the themes of Summit 2016 was “dev and ops coming together through common architectures, processes, and platforms.” This echoes major trends in technology — DevOps and architectures, process, and platform as a unifying IT strategy — and yet none of these concepts are really new. Two decades ago, there were developers and operations, there was enterprise architecture, application platforms, and internal processes. So what’s new and what is bringing the urgency now?

I think the difference comes down to speed (and eventually differences in degree become differences in kind). Twenty years ago, an application was released yearly, sometimes even every couple of years. A patch or security update could take a few months to move in the pipeline from development to testing to production.

Now customers expect patches for security vulnerabilities within hours of them being detected, and the expanding number of applications (from consumer mobile apps to internal systems to IoT devices) means that enterprises have potentially dozens of touchpoints and hundreds of services to maintain.

The “modern” part of modern application development isn’t in the app — it’s in the speed.

This year’s Summit kicked off with three interlocking demos, each showing the different paths and progressions that an IT environment will face as they juggle modernizing existing applications and creating new ones within a heterogeneous (and dynamically changing) ecosystem.

 Lifting and Shifting (Windup)

Continue reading “Summit Notes: Tuesday Morning General Session”

Summit Prep: Containers

Containers, as a technology, bridge two critical aspects of modern application development: process (e.g., DevOps) and platform. This means that containers are a part of pretty much any long-term technology vision because they enable changes in how applications are architected, how teams interact, and how development cycles run.

The containers track at Red Hat Summit covers that range of topics, from the operational side of using containers to the architectural side of working with containers in microservices environments.

Challenges in a microservices age: monitoring, logging, and tracing in OpenShift

Thursday, 4:30, 105

Even new technologies still need to cover the basics of security and maintenance. While emerging technologies and architectures are more nimble and dynamic by being distributed, that increased complexity affects both security and maintenance. This looks at different options within OpenShift for managing enterprise-scale deployments of containers.

Red Hat container technology strategy

Tuesday, 11:30, 153A

This looks at some of the trends and emerging projects related to containers and orchestration.

From monolith to containers: How Verizon containerized legacy applications on OpenShift

Tuesday, 4:30, 153B

Unless you’re a startup, you already have existing infrastructure and apps that are central to your business. No matter how cool or innovative new technologies look, there is always the underlying question — how do you move from here to there? This looks at a proof-of-concept environment at Verizon and what lessons they’ve learned on transitioning from monoliths and legacy apps to containers and microservices.

 

A container-platform approach to digital transformation and DevOps

Thursday, 11:30, 154

Part of the reason that containers are so closely aligned with DevOps, conceptually, is because changing your technology only gets you so far — you really have to change your processes (your culture). This looks at that union of process and platform as central to digital transformation.

 

Summit Prep: Integration

Integration is the multi-syllabic yet succinct way to say “getting a bunch of different things to work together usefully.” Looking at the integration track for Red Hat Summit, integration is the theme, but there are as many different paths as there are things to be integrated. This spans everything from messaging to API management to IoT — basically any system that requires things to communicate.

These are the sessions jumped out at me.

The three pillars that apply agile to integration

Tuesday, 3:30, 151A

One emerging phrase is “agile integration” — basically the ability to integrate things in fast moving environment. This explores three capabilities that are critical for integration to be agile: distribution, containers, and APIs.

Red Hat iPaaS – integration made easy

Tuesday, 4:30, 151B

“iPaaS” is the acronym for “integration platform as a service,” and is essentially shorthand for saying that different Red Hat integration projects are available as native OpenShift images — containerized integration.

Building secure IoT solutions

There are two challenges for IoT that conflict with each other — the need for security and the need for devices and services to communicate with each other. Addressing those challenges requires several different paths of integration, each bringing its own level of security and its ability to integrate different aspects of an environment, including Fuse (integration and messaging), BRMS, and 3scale API management.

Introducing JBoss AMQ 7

AMQ 7 introduces a new, more flexible architecture, including a new router component for automatic traffic routing and rerouting, a new high availability design which doesn’t require a centralized data store, and new messaging protocol support.

Summit Prep: Process Driven Apps

The “process driven” part of process driven applications is fascinating because of how many different things it touches — complex event processing, low code development, DevOps processes. BPM is an engine for completely different and unique application workflows; and it also unites business analysts and developers in a unique way by visualizing or applying business logic within application designs.

There is a separate business automation track at Red Hat Summit, and a scan of the different session titles hints at the diversity of concept and application that falls under “automation.” These are just some of my favorites; check the catalog for the full list.

Transforming a process driven enterprise into a digitally engaged enterprise

Tuesday, 11:30, 157B

This looks at automation as a step beyond being responsive to events and as a point where customers can directly engage with an application (and organization).

A design approach to bridge DevOps and business automation

Tuesday, May 2, 3:30

DevOps is usually aligned with containers because of the need to rapidly and repeatedly spin up identical instances and environments for development, testing, and production. But if BPM is the intersection of business analytics and development, then there needs to be a way for the process and rules design to be as iterative and nimble as a DevOps environment.

End to end BPM

Wednesday, 3:30, 153C

This looks at an integration between Signavio’s modeling tools and JBoss BPM Suite as a business process engine.

Thyssenkrupp: Workforce optimization with business resource planner and OpenShift

Wednesday, 10:15, 153C

One of the core roles of BPM is workforce management, effectively and responsibly allocating people across an enterprise. This is an interesting session because it looks at the real life, practical application of BPM within a large enterprise and how they successfully created a dynamic system tat continuously incorporates changing data and patterns. Who doesn’t love a story with a good ending?

Using real-time data to enable real-time analytics

Tuesday, 10:15, 156C

This looks at in-memory computing and architectures that keep data “at the edge” of the environment for more responsiveness.