Collaborative Research: CI-ADDO-NEW: An Open Memory Array Compiler Framework to Support Devices, Circuits and Systems Research

合作研究：CI-ADDO-NEW：支持设备、电路和系统研究的开放式内存阵列编译器框架

• 批准号: 1205493
• 负责人: Matthew Guthaus
• 金额: $ 52.2万
• 依托单位: University of California-Santa Cruz
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-08-01 至 2017-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1205493&HistoricalAwards=false
• 关键词: Collaborative; Research; CI; ADDO; NEW

The OpenRAM project aims to provide an open-source memory compiler framework for Random-Access Memories (RAMs). Most academic Integrated Circuit (IC) design methodologies are inhibited by the availability of memories since commercially available memory compilers are often black box, not modifiable, and have limited available memory configurations. However, memories are the largest barrier to future scaling and require urgent solutions to lower leakage power consumption, improve reliability in shrinking supply voltages, and incorporate post-CMOS technologies. The OpenRAM compiler will enable research in these areas by providing a completely modifiable, verified, technology independent compiler for single-port and multi-port RAMs and many-port register files. The availability of the OpenRAM will enable research in a variety of IC-related disciplinesincluding: computer architecture and system-on-chip (SOC) design, memory circuit and device research, and computer-aided design (CAD).Computer architects and SOC designers need access to a variety of memory configurations including specialized many-ported register files to prototype system-level implementations. Similarly, memory circuit and device researchers need a framework to prototype new circuits and devices in the context of these many configurations. Last, CAD researchers need a framework to study the optimization and analysis of power and yield of on-chip memories. In addition to the broader impact of enabling future memory system scaling, OpenRAM will also be an important education tool by allowing anyone to synthesize and utilize memory macros in standard-cell design flows and implement custom extensions.

OpenRAM项目旨在为随机存取存储器（RAM）提供一个开源的内存编译器框架。大多数学术集成电路（IC）设计方法受到存储器可用性的抑制，因为商业上可用的存储器编译器通常是黑盒，不可修改，并且具有有限的可用存储器配置。然而，存储器是未来扩展的最大障碍，迫切需要解决方案来降低泄漏功耗，提高可靠性，缩小电源电压，并采用后CMOS技术。 OpenRAM编译器将通过为单端口和多端口RAM以及多端口寄存器文件提供完全可修改、可验证、技术独立的编译器来实现这些领域的研究。 OpenRAM的可用性将使各种IC相关学科的研究成为可能，包括：计算机架构和片上系统（SOC）设计、存储器电路和器件研究以及计算机辅助设计（CAD）。计算机架构师和SOC设计人员需要访问各种存储器配置，包括专用多端口寄存器文件，以原型系统级实现。 类似地，存储器电路和器件研究人员需要一个框架来在这些许多配置的背景下对新电路和器件进行原型设计。 最后，CAD研究人员需要一个框架来研究片上存储器的功耗和成品率的优化和分析。除了支持未来内存系统扩展的广泛影响外，OpenRAM还将成为一个重要的教育工具，允许任何人在标准单元设计流程中合成和利用内存宏，并实现自定义扩展。