Bridging the Frequency Gap in Heterogeneous 3D SoCs through Technology-Specific NoC Router Architectures

Bridging the Frequency Gap in Heterogeneous 3D SoCs through Technology-Specific NoC Router Architectures
复制标题

通过特定技术的 NoC 路由器架构弥合异构 3D SoC 中的频率差距

DOI:
10.1145/3394885.3431421
复制
发表时间:
2021
期刊:
2021 26th Asia and South Pacific Design Automation Conference (ASP-DAC)
影响因子:
--
通讯作者:
Thilo Pionteck
Thilo Pionteck
中科院分区:
--
文献类型:
--
作者:
J. M. Joseph;L. Bamberg;J. Geonhwa;Ruei-Ting Chien;Rainer Leupers;Alberto García-Oritz;Tushar Krishna;Thilo Pionteck

文献摘要

参考文献

被引文献

相似文献

在异构型3D片上系统(SoCs)中,性能一致的片上系统(NoC)受到一个主要限制:路由器的时钟频率因制造工艺的不同而不同。例如,数字节点允许路由器的时钟频率高于混合信号节点。这种大的频差通常由复杂且昂贵的伪中同步或异步路由器体系结构来解决。这里,选择了一种更有效的方法来弥合频差。我们建议使用异类网络体系结构。我们表明,减少VC的数量可以弥补高达2倍的频差。在将15 nm数字和30 nm混合信号节点相结合的示例性设置中,与同构同步网络体系结构相比,我们实现了统一随机流量的系统级延迟改进高达47%,秒基准的系统级延迟改进高达59%,最大吞吐量增加50%,面积减少高达68%,功耗降低38%。与异步和伪中同步路由器体系结构相比,提出的优化在面积、功耗方面始终表现更好,平均微片延迟改进可以超过51%。
In heterogeneous 3D System-on-Chips (SoCs), NoCs with uniform properties suffer one major limitation; the clock frequency of routers varies due to different manufacturing technologies. For example, digital nodes allow for a higher clock frequency of routers than mixed-signal nodes. This large frequency gap is commonly tackled by complex and expensive pseudo-mesochronous or asynchronous router architectures. Here, a more efficient approach is chosen to bridge the frequency gap. We propose to use a heterogeneous network architecture. We show that reducing the number of VCs allows to bridge a frequency gap of up to 2x. We achieve a system-level latency improvement of up to 47% for uniform random traffic and up to 59% for PARSEC benchmarks, a maximum throughput increase of 50%, up to 68% reduced area and 38% reduced power in an exemplary setting combining 15-nm digital and 30-nm mixed-signal nodes and comparing against a homogeneous synchronous network architecture. Versus asynchronous and pseudo-mesochronous router architectures, the proposed optimization consistently performs better in area, in power and the average flit latency improvement can be larger than 51%.
DOI: 10.1109/hpca47549.2020.00044
发表时间: 2020-02
期刊: 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA)
影响因子: --
作者:
Mayank Parasar;Hossein Farrokhbakht;Natalie D. Enright Jerger;Paul V. Gratz;T. Krishna;Joshua San Miguel
通讯作者: Mayank Parasar;Hossein Farrokhbakht;Natalie D. Enright Jerger;Paul V. Gratz;T. Krishna;Joshua San Miguel
DOI: 10.1109/vlsisoc.2011.6081643
发表时间: 2011
期刊: 2011 IEEE/IFIP 19th International Conference on VLSI and System-on-Chip
影响因子: --
作者:
P. Vivet;D. Dutoit;Y. Thonnart;F. Clermidy
通讯作者: F. Clermidy
14nm FinFET 技术中的单片 3D IC 与基于 TSV 的 3D IC
DOI: 10.1109/s3s.2016.7804405
发表时间: 2016
期刊: 2016 IEEE SOI-3D-Subthreshold Microelectronics Technology Unified Conference (S3S)
影响因子: --
作者:
S. Samal;D. Nayak;M. Ichihashi;S. Banna;S. Lim
通讯作者: S. Lim
在 Monolithic 3D 中设计垂直处理器
DOI: 10.1145/3307650.3322233
发表时间: 2019
期刊: 2019 ACM/IEEE 46th Annual International Symposium on Computer Architecture (ISCA)
影响因子: --
作者:
Bhargava Gopireddy;J. Torrellas
通讯作者: J. Torrellas
ViChaR:用于 NoC 路由器的动态虚拟通道调节器 [39]
DOI: 10.1007/978-90-481-3031-3_3
发表时间: 2009
期刊: 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA)
影响因子: --
作者:
C. Nicopoulos;V. Narayanan;C. Das
通讯作者: C. Das