Matt Grange

Scalability of network-on-chip communication architecture for 3-D meshes

Design Constraints imposed by global interconnect delays as well as limitations in integration of disparate technologies make 3-D chip stacks an enticing technology solution for massively integrated electronic systems. The scarcity of vertical interconnects however imposes special constraints on the design of the communication architecture. This article examines the performance and scalability of different communication topologies for 3-D Network-on-Chips (NoC) using Through-Silicon-Vias (TSV) for inter-die connectivity. Cycle accurate RTL-level simulations are conducted for two communication schemes based on a 7-port switch and a centrally arbitrated vertical bus using different traffic patterns. The scalability of the 3-D NoC is examined under both communication architectures and compared to 2-D NoC structures in terms of throughput and latency in order to quantify the variation of network performance with the number of nodes and derive key design guidelines.

On signalling over through-silicon via (TSV) interconnects in 3-D integrated circuits

This paper discusses signal integrity (SI) issues and signalling techniques for Through Silicon Via (TSV) interconnects in 3-D Integrated Circuits (ICs). Field-solver extracted parasitics of TSVs have been employed in Spice simulations to investigate the effect of each parasitic component on performance metrics such as delay and crosstalk and identify a reduced-order electrical model that captures all relevant effects. We show that in dense TSV structures voltage-mode (VM) signalling does not lend itself to achieving high data-rates, and that current-mode (CM) signalling is more effective for high throughput signalling as well as jitter reduction. Data rates, energy consumption and coupled noise for the different signalling modes are extracted.

A scalable multi-dimensional NoC simulation model for diverse spatio-temporal traffic patterns

This paper describes a powerful simulation platform that enables accurate simulations of numerous network configurations under realistic traffic patterns to predict the performance and power needs of a 3-D integrated system early in the design flow. The simulation platform can model virtually any sized 2-D or 3-D network configuration, providing low-cost and fast tradeoff evaluations of various systems architectures. The network simulator uses scalable RTL-level models that can be used for accurate power and timing analyses. We demonstrate the capability of our simulation model by analyzing the performance of various network topologies under spatio-temporal traffic patterns to show how the network topology can be adjusted to meet the performance requirements of a design before it is manufactured. The simulation results can be used to optimize the placement of cores and communication buses early in the flow. By using the model, standard applications such as mobile application processor, femto-cell base-stations on-chip and wide-IO TSV memory stacking can be simulated.

Optimal signaling techniques for Through Silicon Vias in 3-D integrated circuit packages

We investigate optimal techniques for signaling over Through Silicon Vias (TSV) in 3-D circuits to derive design guidelines for maximizing data rate, energy and signal integrity. Low-voltage differential (LVDS) and low-voltage single-ended (LVSE), voltage mode (VM) and current mode (CM) drivers and receivers are implemented in a 65 nm CMOS technology and SPICE simulations with accurate TSV electrical models including coupling effects extracted from a commercial field solver. Trade-offs between the signaling circuits are discussed and the results quantified in terms of data rate, delay variation, noise amplitude and energy consumption.

Physical mapping and performance study of a multi-clock 3-Dimensional Network-on-Chip mesh

The physical performance of a 3-Dimensional Network-on-Chip (NoC) mesh architecture employing Through Silicon Vias (TSV) for vertical connectivity is investigated with a cycle-accurate RTL simulator. The physical latency and area impact of TSVs, switches, and the on-chip interconnect is evaluated to extract the maximum signaling speeds through the switches, horizontal and vertical network links. The relatively low parasitics of TSVs compared to the on-chip 2-D interconnect allow for higher signaling speeds between chip layers. The system-level impact on overall network performance as a result of clocking vertical packets at a higher rate through the TSV interconnect is simulated and reported.

Optimal network architectures for minimizing average distance in k-ary n-dimensional mesh networks

A general expression for the average distance for meshes of any dimension and radix, including unequal radices in different dimensions, valid for any traffic pattern under zero-load condition is formulated rigorously to allow its calculation without network-level simulations. The average distance expression is solved analytically for uniform random traffic and for a set of local random traffic patterns. Hot spot traffic patterns are also considered and the formula is empirically validated by cycle true simulations for uniform random, local, and hot spot traffic. Moreover, a methodology to attain closed-form solutions for other traffic patterns is detailed. Furthermore, the model is applied to guide design decisions. Specifically, we show that the model can predict the optimal 3-D topology for uniform and local traffic patterns. It can also predict the optimal placement of hot spots in the network. The fidelity of the approach in suggesting the correct design choices even for loaded and congested networks is surprising. For those cases we studied empirically it is 100%.