{"title":"ERAsim: A Flexible Python-based Architectural Modeling and Simulation Framework","authors":"Saud Wasly","doi":"10.1109/ICIAS49414.2021.9642593","DOIUrl":null,"url":null,"abstract":"Full-System simulators remain essential tools to quickly design and test novel architectural ideas. This work introduces a system-level hardware modeling and simulation framework with an emphasis on ease of use. It is entirely based on Python to improve developer productivity. At the same time, meta-tracing JIT compilation techniques are used to improve simulation speed without losing timing accuracy. The proposed framework supports modeling components ranging from cycle-accurate to purely functional. The simulation engine combines time-driven simulation with transaction-level modeling to allow for precise timing estimates without compromising simulation speed. The engine also supports multi-threaded execution for multi-master systems.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICIAS49414.2021.9642593","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Full-System simulators remain essential tools to quickly design and test novel architectural ideas. This work introduces a system-level hardware modeling and simulation framework with an emphasis on ease of use. It is entirely based on Python to improve developer productivity. At the same time, meta-tracing JIT compilation techniques are used to improve simulation speed without losing timing accuracy. The proposed framework supports modeling components ranging from cycle-accurate to purely functional. The simulation engine combines time-driven simulation with transaction-level modeling to allow for precise timing estimates without compromising simulation speed. The engine also supports multi-threaded execution for multi-master systems.