Abstract

Heterogeneous 3D System-on-Chips (3D SoCs) are the most promising design paradigm to combine sensing and computing within a single chip. A special characteristic of communication networks in heterogeneous 3D SoCs is the varying latency and throughput in each layer. As shown in this work, this variance drastically degrades the network performance. We contribute a co-design of routing algorithms and router microarchitecture that allows to overcome these performance limitations. We analyze the challenges of heterogeneity: Technology-aware models are proposed for communication and thereby identify layers in which packets are transmitted slower. The communication models are precise for latency and throughput under zero load. The technology model has an area error and a timing error of less than 7.4% for various commercial technologies from 90 to 28nm. Second, we demonstrate how to overcome limitations of heterogeneity by proposing two novel routing algorithms called Z+(XY)Z- and ZXYZ that enhance latency by up to 6.5x compared to conventional dimension order routing. Furthermore, we propose a high vertical throughput router microarchitecture that is adjusted to the routing algorithms and that fully overcomes the limitations of slower layers. We achieve an increased throughput of 2 to 4× compared to a conventional router. Thereby, the dynamic power of routers is reduced by up to 41.1% and we achieve improved flit latency of up to 2.26× at small total router area costs between 2.1% and 10.4% for realistic technologies and application scenarios.

Highlights

  • 3D integration is one of the most promising paradigms to meet the perpetual demand for chips with higher performance, less power consumption and reduced area [1]

  • To the best of our knowledge, there are no related works which consider the relationship between routing algorithms and architectures but this topic is very relevant in heterogeneous 3D System-on-Chips (3D SoCs) due to latency and throughput limitations

  • MODELING NoC COMMUNICATION IN HETEROGENEOUS 3D SoCs We model the horizontal and vertical communication separately, since different factors are relevant: Communication within a layer is synchronous while communication between layers is not always

Read more

Summary

RELATED WORK

As already stated in the introduction, we discuss here why existing 3D interconnects are not considering heterogeneity sufficiently. The majority of routing algorithms for 3D interconnects do not consider performance differences between routers due to varying technology nodes; yet, this effect is significant as we will show in this work. To the best of our knowledge, there are no related works which consider the relationship between routing algorithms and architectures but this topic is very relevant in heterogeneous 3D SoCs due to latency and throughput limitations. 5) We model synchronous routers within layers and not purely synchronous routers between layers, following a GALS approach (globally asynchronous, locally synchronous) This is reasoned as follows: Heterogeneous 3D interconnects will be in non-purely synchronous settings, since components in disparate technologies are potentially clocked at varying speeds and the slowest, synchronous clock wastes performance. The effect of both factors is encapsulated into an abstract model It covers the influence of technology nodes, constraints of synthesis tools and router architectures. Throughput and transmission speed under zero load

HORIZONTAL COMMUNICATION
TACKLING LATENCY LIMITATIONS VIA NOVEL ROUTING STRATEGIES
TACKLING THROUGHPUT LIMITATIONS VIA NOVEL
TACKLING LATENCY VIA ROUTING STRATEGIES
APPLYING PRINCIPLE 1
APPLYING PRINCIPLE 2
PROOF OF VALIDITY
VIII. RESULTS
Findings
DISCUSSION
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call