Process-Based Simulation with Stackless Coroutines

Overview

This repository contains the source code for the simulator core discussed in our SAM 2020 paper. In order to run the examples, you should install the Rust-compiler first.

The src/ directory contains the source code for our simulator core simcore-rs as well as the decompressor-tokenizer- and the barbershop-example from the paper. The folder benches/ contains the code used for benchmarking the barbershop- example. In slx/ one can find the SLX-code for the barbershop-example, including benchmarking provisions. The results/ directory contains our benchmarking results for the barbershop-example in Rust and SLX.

Unfortunately, we are unable to provide a link to the SLX-compiler at this time. Please contact us if you wish to execute the barbershop-example in SLX as well.

Executing the Sample Code

You may execute the decompressor-tokenizer example using cargo run --bin coroutines and the barbershop example using cargo run --bin barbershop. For the benchmarks of the simulator core in case of the barbershop example you may enter cargo bench into your terminal. Cargo will download and compile all the dependencies for you.

Should you wish to execute the SLX code as well, you will need to obtain a version of SLX first. The free student version is sufficient to run the benchmarks with a sample size of 1 for a quick comparison between SLX and Rust, but in order to get statistically relevant results, you need to own a license for the full version of SLX.

Benchmarks

We executed the benchmarks under Windows 10 on an Intel Core i7-10750H processor with 6 cores and 32 GB DDR4 RAM. You can find the results under results/barbershop.lis for SLX and results/barbershop/report/index.html for Rust. In a nutshell: we see a speed difference between the SLX- and the Rust-version of the barbershop example of about 3 in favor of SLX. We have summarized the results in the following table for your convenience: