NServiceBus 3.0 Released
I’m thrilled to announce the release of NServiceBus 3.0.
This is the biggest, most significant release we’ve done by far. This version has been in development since our release of version 2.0 almost exactly 2 years ago, and is coming 6 months after our version 2.6 release.
During this time we’ve seen NServiceBus be adopted by the world’s largest financial institutions, top retail brands, leading airlines and transportation organizations, international insurance conglomerates, global healthcare providers, education institutions, government bodies, and other companies in energy, media, and online gambling (to name a few).
What makes 3.0 special
It’s not just the new features visible on the surface (which we have a lot of), it’s the inner architecture.
I know that as a creator I will always have deeper appreciation for its internals, but it’s this architecture which enabled us to take our already impressive extensibility and go an order of magnitude further. In fact, our ability to take what was originally designed to be an entirely on-premise technology and run that entirely in the cloud (and as a hybrid), is a result of this architecture.
Azure Cloud Support
From support for web and worker roles, to Azure storage queues and AppFabric queues, and SQL Azure – NServiceBus stitches together all of these technologies to give you the most reliable combination of all of these technologies, all with the simple coding and convention-over-configuration defaults developers have come to love in NServiceBus.
The new DataBus capability in version 3.0 allows you to go beyond the 8-256KB message size limits of Azure, and also the 4MB limits of MSMQ.
In short, you can target the NServiceBus API and keep your code clean of whether you’re deploying on premise or to the cloud.
For Admins and the Business
In version 2.6 we provided performance counters telling administrators not only how long it took to process messages, but also included the time those messages waited in the queue giving a much clearer picture of the system’s ability to handle the production load. In 3.0, we’re taking this to the next level.
You specify on an endpoint what is its Service Level Agreement (SLA) and at runtime NServiceBus will look at how the processing and wait times are trending – telling you just how soon an endpoint will break that SLA (assuming load continues the same way). This enables administrators to focus on those parts of the system that need to be scaled right away, even if they were at a lower absolute percent of SLA than others.
Business Service Level Monitoring – done right.
Smarter conventions coupled with a bundled deployment story give developers the power of scaling out, cross-datacenter communication, and durable timeout management with almost no configuration at all. Our use of RavenDB takes the NServiceBus deployment story and makes it friction-free as well. You now get all the power tools in NServiceBus and they “just work”, just like everything else.
But I’d have to say my favorite feature is the new Deferred Messaging exposed on the bus for simple messaging as well as on sagas for long-running processes. The ability to take some data that comes into your system and say “this is interesting data, but I’d like to handle it 2 days from now” and just like that toss it into the future – well, it twists my brain into wonderful new insights about the business.
And that only scratches the surface – for more information click here.
Having crossed the 50,000 download mark, and with an active developer community of around 1500, NServiceBus is becoming a platform in its own right. And we’re far from done – there’s so much more we’ve got planned.
And for those of you using version 2.6, we’ve put together a page to take you through the upgrade step-by-step – including a wonderful video from our own Andreas Ohlund. Take a look.
And just wait until you see the things we’ve got coming.