Abstract:With the continuous reduction in transistor size, the power consumption issue associated with the adoption of Network-on-Chips (NoCs) in multicore systems has become increasingly prominent and severe, particularly in terms of static power consumption. Currently, the techniques employed to address the issue of static power consumption in NoCs can be broadly categorized into two forms: power gating and simplified router architecture. However, these designs face critical bottlenecks in maintaining network performance stability and flexibility. In this paper, we propose a power gating approach utilizing double buffering to conserve power. Firstly, we introduce the DBU-PG router architecture, which effectively reduces network power consumption by sharing an input buffer between every two input ports when the network load is low. Secondly, through improved arbitration and flow control mechanisms, packets can be efficiently transmitted between shared buffers. Finally, we propose an improved router architecture, DBU-PG2, which enables packets in different virtual channels of the same buffer to be executed concurrently. Based on evaluation in real application, our DBU-PG and DBU-PG2 routers achieved reductions of 69% and 66%, respectively, in static power consumption compared to the baseline. Additionally, DBU-PG2 reduced packet latency by 7.5% while incurring only a 2.3% increase in area overhead.

Optimizing Buffer Usage For Networks-On-Chip Design

Buffer Planning for Application-Specific Networks-on-chip Design.

A multi-path routing method with shared buffer on NoC

A High-Throughput and Size-Efficient NoC Buffer Design Method

Evaluation of buffer organizations for network-on-chip

An adaptive buffer-shared routing algorithm for NoC designs

An adaptive buffer regulating scheme for on-chip routers

Early‐stage Microfluidic Network Design Framework Using Graph Sparsificiation Based Optimisation

A Tightly Coupled Network-on-Chip Router Architecture

Self-optimizing Two-layer Network-on-Chip Based on Dominant Network-Flow Adaption

Achieving balanced buffer utilization with a proper co-design of flow control and routing algorithm

Adaptive Congestion Control for Application Specific Networks-On-Chip

Pareto-Optimization Framework for Automated Network-on-Chip Design

DBU-PG: energy-efficient noc design using dual-buffering power gating

Interactive buffer sizing for core routers

Pull-off Buffer: Borrowing Cache Space to Avoid Deadlock for Fault-Tolerant NoC Routing

A Mathematical Method for Analyzing the Latency and Throughput of Mesh with Buffer Optimization

A Power-Aware Adaptive Routing Scheme for Network on a Chip

A Buffer Planning Algorithm with Congestion Optimization.

Buffer Sizing for Core Routers from the Perspective of the Network Layer

Design Trade-Offs in Packetizing Mechanism for Network-on-Chip