Heap architectures for concurrent languages using message passing

We discuss alternative heap architectures for languages that rely on automatic memory management and implement concurrency through asynchronous message passing. We describe how interprocess communication and garbage collection happens in each architecture, and extensively discuss the tradeoffs that axe involved. In an implementation setting (the Erlang/OTP system) where the rest of the runtime system is unchanged, we present a detailed experimental comparison between these architectures using both synthetic programs and large commercial products as benchmarks.