Akka, formerly Lightbend, today at the KubeCon + CloudNativeCon 2024 conference, unfurled a revamped Akka that adds a software development kit (SDK) to make the platform-as-a-service (PaaS) environment for building distributed Java applications more accessible.
In addition, Akka is making available a serverless instance of the PaaS, dubbed Akka 3, along with enabling IT teams to deploy the framework on the cloud computing environment of their choice.
Akka 3 also now makes it possible to replicate applications across clouds, regions and devices to enable no-downtime migrations, repatriation to on-premises IT environments, and disaster recovery.
Company CEO Tyler Jewell said Akka is now the first application runtime to provide multi-master replication, which is made possible because Akka applications have access to their own in-memory databases with writable conflict-free replicated data types (CRDT) running globally.
Finally, Akka is now indemnifying customers for any losses stemming from data that becomes unrecoverable. Akka can make that pledge because each application has its own in-memory database that makes it unlikely data will ever be lost, said Jewell.
More than 100,000 applications have already been built using Akka and the decision to rename Lightbend was made to align the company more closely with an Akka framework that has been downloaded one billion times, noted Jewell. The overall goal is to make it possible for IT organizations to build and deploy truly elastic applications that consume infrastructure resources more efficiently.
IT teams are adopting Akka because it enables them to guarantee service level agreements (SLAs) by running applications that can continuously adapt independently of IT infrastructure environments, said Jewell. Every other framework that might be used to build these types of applications can’t guarantee that data won’t be lost, said Jewell.
Akka 3 is the culmination of an eight-year effort to turn a set of libraries into a more opinionated PaaS environment that an SDK now makes it easier to build high-performance distributed Java applications capable of running at scale, he added.
It’s not clear how large the Akka developer community is. However, organizations of all sizes are increasingly focused on ensuring their applications are portable in case, for example, licensing terms are altered in a way that suddenly increases costs. At the same time, more data than ever needs to be processed in memory to support applications running in near real-time. Much of the data needs to be processed and analyzed in memory at the point where it is either collected or created. The challenge is that data eventually needs to be synchronized with some type of central repository that typically resides in the cloud or an on-premises IT environment.
Regardless of whether IT teams adopt a publish-subscribe approach to building these types of applications or some other type of event-driven programming model, these applications are much more challenging to build and maintain than traditional batch-oriented applications. As such, they will no doubt require a level of DevOps expertise to build, deploy and maintain across highly distributed computing environments that depend on network connectivity that is not always reliable.
