Pushing Enterprise Software to the Next Level - Self-contained Web Applications on In-Memory Platforms

Technologies covered:
Non-volatile memory, self-contained systems architecture, and Starcounter

Purpose of the talk
Random-access memory is getting cheaper, web technologies are mature and architectures for true modularity are well established. What happens if we use these three facts together?

In this talk, I’ll introduce a new type of architecture where an application server and a database share memory. Thus, data objects are only stored once, and overhead communication between the two tiers is eliminated. Effectively, an application’s heap is altered with durable software transactional memory, while the physical database that the application accesses is virtualized to match its conceptual data model. As a result, it gives us higher performance, a simplified stack, and increased modularity.

I will focus on the self-contained systems approach utilized on the in-memory platform. This architecture focuses on a separation of the functionality into many independent systems. A complete logical system is a collaboration of many smaller, fully functional, domain oriented applications.
 

Target Audience
This is targeted to the technical audience, from developers to architects and CTOs. The audience is someone who is looking for a new way to utilize in-memory computing to build modular, enterprise web-applications instead of using it only as a caching layer. In this talk I will use Starcounter’s in-memory application platform to visualize examples of this technology.

The Take-Away
After attending this talk, the audience will have broaden their knowledge of how in-memory computing changes the way we build apps. They will see real use cases build on a new technology. I aim to inspire the audience on how to cut developing time and achieve a faster time-to-market.