OpenABL: A Domain-Specific Language for Parallel Agent-Based Simulations

OpenABL is a domain-specific language for portable, high-performance, parallel agent modeling. The artifact includes source code, build instructions, and benchmark files for six sample applications from different fields. It was developed by Biagio Cosenza to support reproducible simulations across different parallel implementations and hardware platforms.

Use Cases

• Benchmarking performance of agent-based simulation toolkits like Flame and Mason based on the six included sample applications.
• Developing portable agent-based models that can target different computing architectures such as CPUs, GPUs, and clusters.
• Reproducing simulation results from the related Euro-Par 2018 publication using the provided code and input files.

Strengths

• Includes six experimental agent-based models from different domains for benchmarking.
• Designed to compile to multiple backends targeting single CPUs, GPUs, and clusters.
• Artifact contains source code, benchmark scripts, and example input files for reproducibility.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Biagio Cosenza

Collection Method

Artifact includes source code, benchmarks, and examples related to the Euro-Par 2018 paper.