PASTA：一种用于加速器的模块化程序分析工具框架
PASTA: A Modular Program Analysis Tool Framework for Accelerators
AI总结：
PASTA是一种低开销、模块化的程序分析工具框架，适用于加速器环境，提供高效的性能分析和工具扩展能力。
摘要：
现代计算系统中硬件加速器的复杂性和多样性日益增加，要求灵活且低开销的程序分析工具。我们提出了PASTA，一种低开销且模块化的用于加速器的程序分析工具框架。PASTA抽象了低级剖析API和多种深度学习框架，为用户提供统一的接口，以在多个层次上捕获和分析运行时事件。其可扩展设计使研究人员和实践者能够快速原型设计定制工具，且开销极低。我们通过开发几种分析工具来展示PASTA的实用性，包括一个深度学习工作负载特征化工具和一个UVM优化工具。通过在NVIDIA和AMD GPU上对主流深度学习工作负载进行广泛评估，包括单GPU和多GPU场景，我们展示了PASTA的广泛适用性。在NVIDIA GPU上，我们进一步表明，PASTA提供了详细的性能洞察，其开销显著低于传统分析工具，比传统分析工具快高达1.3*10^4倍，这归功于其GPU加速的后端。PASTA在实用性、可扩展性和效率之间取得了良好的平衡，使其适合现代基于加速器的计算环境。
英文摘要：
The increasing complexity and diversity of hardware accelerators in modern computing systems demand flexible, low-overhead program analysis tools. We present PASTA, a low-overhead and modular Program AnalysiS Tool Framework for Accelerators. PASTA abstracts over low-level profiling APIs and diverse deep learning frameworks, offering users a unified interface to capture and analyze runtime events at multiple levels. Its extensible design enables researchers and practitioners to rapidly prototype custom tools with minimal overhead. We demonstrate the utility of PASTA by developing several analysis tools, including a deep learning workload characterization tool and a UVM optimization tool. Through extensive evaluation on mainstream deep learning workloads tested on NVIDIA and AMD GPUs under both single- and multi-GPU scenarios, we demonstrate PASTA's broad applicability. On NVIDIA GPUs, we further show that PASTA provides detailed performance insights with significantly lower overhead, up to 1.3*10^4 faster than conventional analysis tools, thanks to its GPU-accelerated backend. PASTA strikes a practical balance between usability, extensibility, and efficiency, making it well-suited for modern accelerator-based computing environments.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
