Modeling and Simulation Frameworks for Processing-in-Memory Architectures

2025-11-26

AI Summary

To address the lack of systematic evaluation methodologies for Processing-in-Memory (PIM) architectures, this work proposes a multi-level, cross-abstraction modeling and simulation framework. The framework introduces the first three-dimensional taxonomy of PIM simulation approaches—categorized by abstraction level, design objectives, and evaluation metrics—and integrates models spanning device-level to application-level abstractions. It supports mainstream PIM technologies including HBM-PIM and ReRAM, and incorporates low-latency memory access simulation, compute-memory co-modeling, and standardized benchmark suites. Through a comprehensive survey of existing tools, we identify fundamental trade-offs among fidelity, scalability, and compatibility, and clarify open challenges. Our framework significantly improves the accuracy of simulation tool selection and enhances result comparability across studies. It establishes both theoretical foundations and practical guidelines for developing next-generation PIM modeling tools that are high-fidelity, reusable, and extensible.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Hardware-aware ML
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Processing-in-Memory (PIM) has emerged as a promising computing paradigm to address the memory wall and the fundamental bottleneck of the von Neumann architecture by reducing costly data movement between memory and processing units. As with any engineering challenge, identifying the most effective solutions requires thorough exploration of diverse architectural proposals, device technologies, and application domains. In this context, simulation plays a critical role in enabling researchers to evaluate, compare, and refine PIM designs prior to fabrication. Over the past decade, a variety of PIM simulators have been introduced, spanning low-level device models, architectural frameworks, and application-oriented environments. These tools differ significantly in fidelity, scalability, supported memory/compute technologies, and benchmark compatibility. Understanding these trade-offs is essential for researchers to select appropriate simulators that accurately map and validate their research efforts. This chapter provides a comprehensive overview of PIM simulation methodologies and tools. We categorize simulators according to abstraction levels, design objectives, and evaluation metrics, highlighting representative examples. To improve accessibility, some content may appear in multiple contexts to guide readers with different backgrounds. We also survey benchmark suites commonly employed in PIM studies and discuss open challenges in simulation methodology, paving the way for more reliable, scalable, and efficient PIM modeling.

Problem

• Develop frameworks to simulate Processing-in-Memory architectures
• Evaluate diverse PIM designs before physical fabrication
• Guide researchers in selecting appropriate PIM simulation tools

Innovation

• Simulation tools evaluate PIM designs before fabrication
• Categorizes simulators by abstraction levels and objectives
• Surveys benchmark suites and discusses simulation challenges