现场可编程门阵列
计算机科学
计算机体系结构
延迟(音频)
门阵列
嵌入式系统
互连
可编程阵列逻辑
可编程逻辑阵列
布线(电子设计自动化)
组分(热力学)
逻辑综合
逻辑门
逻辑族
电信
物理
算法
热力学
作者
Andrew Boutros,Vaughn Betz
标识
DOI:10.1007/978-981-15-6401-7_49-1
摘要
Since their inception more than thirty years ago, field-programmable gate arrays (FPGAs) have grown more complex, more capable, and more diverse in their applications. FPGAs can be reprogrammed at a fundamental level, changing the function and interconnection of millions of elements. By reconfiguring their hardware to match the application, FPGAs often achieve higher energy efficiency, lower latency or faster time-to-market across a very wide range of application domains. A modern FPGA combines many components, from logic blocks, programmable routing and memory blocks to networks-on-chip and processor subsystems. For best efficiency, each component must be carefully architected to match the needs of a wide range of applications, and to mesh well with the other components. Their design involves many different choices starting from the high-level architectural parameters down to the transistor-level implementation details. This chapter describes the evolution of these FPGA components, their design principles and implementation challenges.
科研通智能强力驱动
Strongly Powered by AbleSci AI