Artificial Intelligence (AI) has witnessed remarkable growth, particularly through the proliferation of Deep Neural Networks (DNNs). These powerful models drive technological advancements across various domains. However, to harness their potential in real-world applications, specialized hardware accelerators are essential. This demand has sparked a market for parameterizable AI hardware accelerators offered by different vendors. Manufacturers of AI-integrated products face a critical challenge: selecting an accelerator that aligns with their product's performance requirements. The decision involves choosing the right hardware and configuring a suitable set of parameters. However, comparing different accelerator design alternatives remains a complex task. Often, engineers rely on data sheets, spreadsheet calculations, or slow black-box simulators, which only offer a coarse understanding of the performance characteristics. The Abstract Computer Architecture Description Language (ACADL) is a concise formalization of computer architecture block diagrams, which helps to communicate computer architecture on different abstraction levels and allows for inferring performance characteristics. In this paper, we demonstrate how to use the ACADL to model AI hardware accelerators, use their ACADL description to map DNNs onto them, and explain the timing simulation semantics to gather performance results.

翻译：人工智能（AI）取得了显著发展，尤其是深度神经网络（DNN）的蓬勃兴起。这些强大的模型推动了多个领域的技术进步。然而，要在实际应用中充分发挥其潜力，专用硬件加速器不可或缺。这一需求催生了由不同供应商提供的可参数化AI硬件加速器市场。集成AI产品的制造商面临一项关键挑战：选择与其产品性能要求相匹配的加速器。这一决策涉及选择正确的硬件并配置合适的参数集。然而，比较不同加速器设计方案仍是一项复杂任务。通常，工程师依赖数据手册、电子表格计算或缓慢的黑盒模拟器，这些方法仅能提供对性能特性的粗略理解。抽象计算机体系结构描述语言（ACADL）是一种对计算机体系结构框图进行简洁形式化的语言，有助于在不同抽象层级上沟通计算机体系结构，并支持推断性能特征。本文展示了如何利用ACADL对AI硬件加速器进行建模、通过其ACADL描述将DNN映射到加速器上，并解释了时序仿真语义以收集性能结果的方法。