← 返回 JSSC 论文列表
📄 下载 JSSC 原文 PDF
JSSC 2015第7期Clocking & PLLs40nmPLLProcessor/CPU

Clock and Synchronization Networks for a 3 GHz 64 Bit ARMv8 8-Core SoC

本文介绍了一种用于3 GHz 64位ARMv8 8核SoC的时钟分布与同步网络设计。
40nm CMOS, 3.0 GHz, 0.8 ps/mV rms, 9 ps period jitter and skew
时钟分布同步网络ARMv8SoC动态频率跳变
动态频率跳变技术(DVFS):通过系统PLL实现动态频率切换,支持云计算平台的多任务负载优化,实测rms抖动低至1ps,显著提升能效比。
Star/H-Mesh混合拓扑结构:结合星型与网状拓扑优势,采用CML和CMOS混合电路设计,实现时钟分布网络周期抖动0.8ps/mV和偏移9ps,优化高频信号完整性。
本地占空比调整电路(DCA):每个核心集成Duty Cycle Adjustment模块,通过动态调节时钟占空比缓解关键路径时序压力,实测处理器性能提升5%以上。
增强型锁存器设计:提出新型高可靠性锁存器结构,将MTBF(平均无故障时间)提升5个数量级,适用于3GHz高频同步操作,显著降低同步错误率。
Abstract
This paper describes the clock distribution and syn- chronization network for a 64 bit ARMv8 8-core microprocessor . Embedded in a SoC for cloud com puting platforms, the processor is fabricated in a 40 nm CMOS technology and operates at 3.0 GHz. The system PLL has a measured rms jitter 1p s a n d f e a t u res dynamic frequency hopping for DVFS applications. In conjunc- tion with a Star/H/Mesh topology, the clock distribution uses both CML and CMOS circuits to minimize period jitter and nominally achieves 0.8 ps/mV rms and 9 ps of period jitter and skew, re- spectively. By using local Duty Cycle Adjustment circuits in each core to properly offset the clock duty cycle and e ase timing crit- ical paths, the processor perfor mance improves by more than 5%. A simple probing circuit for high speed clock measurements can be used to monitor the high frequency excu rsions of the internal supply to counteract any timing violation that could occur. Finally an enhanced latch, which improves MTBF by up to 5 orders of magnitude and thus is suited for high sp eed synchronization oper- ations, is proposed.