muRelBench: MicroBenchmarks for Zonotope Domains

2024-04-24

arXiv.org

AI Summary

Existing weakly relational abstract domains—particularly the zonotope domain—lack standardized benchmarks, hindering rigorous evaluation of algorithmic correctness, performance, and reproducibility. Method: We introduce the first lightweight synthetic benchmark suite specifically designed for the zonotope domain, targeting core operations such as domain closure. Grounded in abstract interpretation theory and formalized via precise semantic constraints, our framework automatically generates test programs with controllable complexity and formal semantic guarantees. Contribution/Results: This work establishes the first standardized micro-benchmarks for zonotope analysis, enabling scalable and formally verifiable experimental evaluation. Empirical evaluation demonstrates that our benchmark significantly improves fairness, reproducibility, and diagnostic capability in comparative studies across diverse zonotope-based algorithms—thereby filling a critical gap in the evaluation infrastructure for weakly relational abstract domains.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Knowledge Representation and Reasoning: Automated Reasoning and Theorem Proving
• Application Domains: Security

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Security and Privacy: Large-scale security measurements

Abstract

We present exttt{muRelBench}, a suite of synthetic benchmarks for weakly-relational abstract domains and their operations. For example, the benchmarks can support experimental evaluations of proposed algorithms such as domain closure.

Problem

• Evaluating algorithms for numerical abstract domains
• Quickly prototyping performance improvements in abstract domains
• Ensuring correctness in synthetic benchmark operations

Innovation

• Extensible microbenchmarking framework for abstract domains
• Enables quick prototyping of numerical domain algorithms
• Includes correctness checks for synthetic benchmarks