Communication Middleware (mw::com)

Overview

LoLa/mw::com: A high-level communication middleware providing a partial implementation of the Adaptive AUTOSAR Communication Management specification. Within the Adaptive AUTOSAR ecosystem, it is commonly known as ara::com, reflecting the namespace defined by the standard.

General introduction

mw::com and LoLa are synonyms right now. While the former just expresses the namespace (mw::com = middleware communication), the latter is a hint on its technical implementation/unique selling point (LoLa = Low Latency). Since LoLa/mw::com is our in-house implementation of the AUTOSAR standard, which – according to this standard – resides under ara::com, the namespace similarity (mw::com ↔ ara::com) is intentional.

Low latency in our mw::com implementation is realized by basing our underlying technical binding on zero-copy shared-memory based communication. The notion of a technical binding also comes from the ara::com AUTOSAR standard: ara::com just defines a standardized user-level API and its behavior (which we generally follow with our mw::com implementation). How an implementation realizes the public API within its lower layers (technical binding) is completely up to it. It is even common, that implementations of the ara::com standard come up with several different technical bindings (e.g. one for local and one for network communication).

Documentation for users

If you are an adaptive application developer in the IPNEXT project, and you want to use mw::com to do local interprocess communication, you will find the user documentation here: LINK TODO.

Documentation of mw::coms public API with usage examples can be found in doc/user_facing_API_examples.md. If you want to see an example app, which uses LoLa to establish shared-memory communication between a service and a client you can take a look at our macro benchmark app which resides in performance_benchmarks/macro_benchmark.