Scale Stack aims to integrate common network services used for building scalable architectures into a single project. The goal is to increase the efficiency by sharing both code and resources between these services. It also aims to be both a service stack and a development framework. For example, protocol parsing modules are written generically so they can be used for any number of client, server, or proxy applications. New services can easily be built be connecting together previous modules in new ways. If you are writing network service software in C/C++, Scale Stack may provide just what you need for your “Kernel” and core modules, then your custom service modules can be built from there.

The design is extremely modular and uses dependency tracking between modules for easy bootstrapping of applications. Non-blocking event processing is at the core of the design, as well as emphasis on security, efficiency, and multi-tenancy. All code is open source under the Apache 2.0 license, so it may be used in a variety of environments. It is written in C++ and follows strict coding guidelines to keep the code maintainable.

C++ was chosen for the runtime efficiency over many of the dynamic and byte-code compiled languages. C++ was chosen over C to allow for safe and clean class abstractions for module writers. While this interface could be implemented in C, C++ provides more protection and compile-time safety checks. Not all features of C++ are used due to design decisions and portability issues, please see the README file in the source for further information.

Why not build on existing software?

Who works on it?