基于人工神经网络的NoC智能动态链路管理方法

功耗是片上网络(NoC)主要限制因素,链路状态的选择性开/关切换算法可降低电路级和系统级的链路功耗,这些算法大多集中于一个简单的静态阈值触发机制,该机制决定了是否应该打开或关闭链路;为解决上述触发机制存在诸多限制,提出了一种针对NoC的人工神经网络(ANN,artificial neutral network)作为动态链路功耗管理方法,该方法基于对系统状态的有监督在线学习,通过使用小型可扩展的神经网络来关闭和打开链路,从而提高预测能力;基于人工神经网络的模型利用了非常低的硬件资源,并且可以集成在大型网状和环面NoC中;通过对不同网络拓扑上各种综合流量模型的仿真结果表明,与静态阈值计算相比,该方法在较低的硬件支出下可以节省功耗;可为解决链路管理NoC中的功耗问题提供思路。     

延迟优化的片上网络低功耗映射*

片上网络（NoC）是解决传统基于总线的片上系统（SoC）所面临的功耗、延迟、同步和信号完整性等挑战的有效解决方案。功耗和延迟是NoC设计中的重要约束和性能指标,在设计的各个阶段都存在着优化空间。基于蚁群优化算法,通过通信链路上并发通信事件的均匀分布来降低NoC映射阶段的功耗和延迟。仿真实验表明,与链路通信量负载均衡的方法相比,该方案能进一步在拓扑映射阶段优化功耗和延迟。     

改进量子算法的片上网络映射研究

片上网络（NoC）是解决片上系统（SoC）之间各个IP核通信的主要方法。其中NoC的映射是整个NoC设计过程中最为关键的步骤之一。采用一种改进的方法解决NoC映射问题,该方法基于量子进化算法,并在算法中采用一种改进的更新方法,之后引入精英策略,让所有中间过程的解都参与到迭代中,选择其中最好的解作为每次迭代的NoC映射最终解。使用该方法建立在延时约束下的NoC映射功耗数学模型,实验表明,该方法在NoC映射中能达到降低通信功耗的目的。     

改进模拟退火遗传算法的3D NoC低功耗映射

功耗优化是NoC设计的重要部分,针对将IP (intellectual property)核合理映射NoC的问题,提出一种初始种群优化的模拟退火遗传映射算法.首先以功耗优化为主要目标,通过对初始种群选取方法进行改进来获取功耗更低的映射方案,并针对遗传算法局部最优问题,在遗传算法交叉操作阶段结合模拟退火算法,得到全局最优方案.实验在Windows系统下采用C++语言实现,结果显示,与传统的遗传算法相比,该算法具有较好的收敛性,能快速搜索到较优解,在124个IP核的情况下,采用改进的模拟退火遗传算法进行映射产生的平均功耗比使用遗传算法时降低了32.0%.     

支持双拓扑结构的片上网络评估高层仿真平台

为实现高效的NoC（片上网络）性能评估, 缩短系统芯片的开发周期, 针对时钟精确级的NoC仿真方法进行研究, 提出了一种新型的高层次、高效率仿真平台, 与仅支持网格拓扑结构的传统仿真器相比, 其创新地支持了网格和环型双拓扑结构的性能评估, 同时支持虚通道扩展的路由器结构设计, 能快速得到网络的延迟、吞吐率、功耗等性能结果。实验结果表明, 该仿真平台能准确模拟NoC功能行为, 快速获得其仿真性能, 为NoC设计验证提供了高效的方法。     

基于改进遗传算法的3DNoC低功耗映射研究

二维片上网络（Two-Dimensional Network-on-Chip,2D NoC）在面积、功耗、布局布线、封装密度等方面都已达到了瓶颈。与2D NoC相比,三维片上网络（Three-dimensional Network-on-Chip,3D NoC）有着诸多优势,因此3D NoC逐渐成为一个重要的研究方向。随着3D NoC集成度的提高,低功耗映射逐渐成为研究热点。将贪心算法的思想与遗传算法相结合提出一种改进的遗传算法,用以解决3D NoC低功耗映射问题,相对于传统遗传算法,改进遗传算法具有更优的搜索能力。仿真结果表明,采用改进后的遗传算法解决3D NoC映射问题可以降低功耗,从总体趋势来看随着处理单元数量的增加功耗降低幅度逐渐增大,在120个处理单元情况下总功耗可降低14%。     

片上网络的低功耗自适应数据保护

针对片上网络(NoC)提出了一种低功耗自适应数据保护机制。根据不同的片上网络通信链路错误数目自适应选择，在路由节点之间进行数据保护的跳距，保证了系统芯片功耗效率最优化。实验结果证明，在同样的可靠约束条件下，采用自适应数据保护，其功耗低于节点到节点，端到端的数据保护，特别是对需要高可靠性的NoC通信结构，自适应数据保护机制表现得更为有效。     

片上网络体系结构的研究与进展

片上网络(NoC)是基于多处理器技术的一种新型的计算集成形式，涉及硬件通信结构、中间件、操作系统通信服务、设计方法及工具等。NoC体系结构的设计重点是实现低功耗和高效通信/计算能力。该文介绍了4种新的NoC体系结构，并在同等约束下进行了功耗比较，2D网格结构的功耗最大、性能最差，聚合环面网络结构则最优。     

功耗限制下的NoC测试端口的优化选择方法

提出了一种NoC测试端口位置和数量的优化选取的方法,它在系统功耗限制的条件下,确定input/output端口的对数,以所有核测试路径总和最短为目标,优化选取NoC 测试端口的最佳位置。本方案在测试功耗不超过系统允许的最大功耗条件下,最大限度地选取测试端口的对数来进行并行测试,从而能高效地完成对核的测试,同时又能有效地避免因测试带来的器件损坏。实验结果表明这种方法提高了测试效率,降低了NoC的总体测试代价。     

基于Prim初始种群选取优化遗传算法的三维片上网络低功耗映射

针对将计算任务合理地映射到三维片上网络（NoC）的问题,提出了一种基于遗传算法（GA）的改进算法。GA具有快速随机的搜索能力,Prim算法可在加权连通图内得到最小生成树,改进算法结合了两种算法的优势,将计算任务合理地分配到各个网络节点,对于优化三维片上网络功耗和散热等问题具有很高的效率。通过仿真实验,对所提出的基于Prim算法的改进GA与基本GA的3D NoC映射算法进行了对比,仿真结果显示,基于Prim算法的改进GA平均功耗更低,从总体趋势来看,处理单元数量的增加与功耗降低幅度成正相关,在101个处理单元情况下,平均功耗比基本GA降低32%。     

Bit-accurate energy estimation for Networks-on-Chip

Networks-on-Chip (NoCs) are recognized as the solution to address the communication bottleneck in a Multi-processor System-on-Chip (MPSoC). As NoCs represent a significant part of system consumption, MPSoC designers expect accurate power models in order to produce energy efficient systems. Nowadays, NoC simulators rely on power models that integrate link models without crosstalk modeling. In this study, we present Noxim-XT, a NoC simulator based on Noxim that embeds a link power model with crosstalk modeling. We show that the crosstalk effect has a deep impact on NoC energy consumption since our results demonstrate that classical models generate errors up to 45.5% on the whole NoC energy consumption estimation. In addition, this tool is able to run application-based traffic and we show that under application-based traffics, the energy estimation made by classical models overestimates the NoC energy consumption by up to 50%.     

The 2D digraph-based NoCs: attractive alternatives to?the 2D mesh NoCs

This paper proposes two-dimensional directed graphs (or digraphs for short) as a promising alternative to the popular 2D mesh topology for networks-on-chip (NoCs). Mesh is the most popular topology for the NoCs, mainly due to its suitability for on-chip implementation and low cost. However, the fact that a digraph offers a lower diameter than its equivalent linear array of equal cost motivated us to evaluate digraphs as the underlying topology of NoCs. This paper introduces a family of NoC topologies based on three well-known digraphs, namely de Bruijn, shuffle-exchange, and Kautz. We study topological properties of the proposed topologies. We show that the proposed digraph-based topologies have several attractive features including constant node degree, low diameter and cost, and low zero load latency which result in superior performance over the mesh. We introduce a deadlock-free routing algorithm for the proposed NoC topologies and compare NoCs employing the proposed topologies and the mesh topology in terms of power consumption and performance. Simulation results also reveal that the proposed NoC topologies offer higher performance and consume lower power than the mesh NoC.     

Application-Specific Network-on-Chip synthesis with flexible router Placement

Network-on-Chip (NoC) has been proposed as a possible solution to the communication problem in nanoscale System-on-Chip (SoC) design. NoC architectures with optimized application-specific topologies have been found to be superior to the regular architectures in designing Multi-Processor System-on-Chip (MPSoC) solutions. The application specific NoC design problem takes as input the chip floorplan, library of NoC components, and communication requirements between the tasks of the application. It outputs the positions of the routers in the floorplan, such that, all communication requirements of the application are satisfied. This paper presents an Integer Linear Programming formulation of the problem, followed by a heuristic technique based on Particle Swarm Optimization (PSO) for finding the router positions from the set of available positions within the chip floorplan. The goal is to minimize the communication cost between cores, satisfying both the link length and router port constraints. The results have been shown on realistic benchmarks. Comparisons have been carried out with regular mesh and custom architectures having routers positioned at (i) the corners of the cores, (ii) the centers of the cores, and (iii) the intersections of the cores. Significant reductions in communication cost have been observed over all the cases. For smaller benchmarks, the optimum results obtained via ILP matches exactly with those reported by the PSO. Many of the existing router placement policies fail even for these small benchmarks, when restrictions are imposed on permissible link length. This establishes the merit of the PSO formulation. Link and router energy consumption of the synthesized NoC have been compared with regular mesh based architectures. The results show significant reduction in communication cost, area overhead, link energy and router energy in the synthesized NoC over regular mesh topology as well.     

基于云自适应遗传算法的NoC映射研究

NoC映射是NoC设计中的重要步骤,映射结果的优劣对NoC的QoS约束和通信功耗有着很大的影响。提出一种采用云自适应遗传算法实现NoC映射的方案,该算法利用云模型对传统遗传算法加以改进,以此新方法自动调整遗传算法过程中的交叉概率和变异概率,从而达到优化遗传算法的目的。结合NoC映射中的具体问题,在功耗和延时约束的限制条件下,建立了延时约束下的NoC映射功耗数学模型。实验表明,该方法在NoC映射中取得了良好的效果,降低了通信功耗。     

Application-driven dynamic bandwidth allocation for two-layer network-on-chip design

Network-on-Chip (NoC) architecture has been widely used in many multi-core system designs. To improve the communication efficiency and the bandwidth utilization of NoC for various applications, we firstly propose a table-based algorithm for identifying the dominant flows at runtime. Then a two-layer NoC architecture with an application-driven bandwidth allocation scheme is presented, which is capable of identifying heavy-load dataflows and dynamically reconfiguring point-to-point (P2P) connections to optimize the heavy-load traffic. Experimental results reveal that our design (8 × 8 mesh NoC) achieves 28.5% performance improvement and 25.9% power consumption saving compared to the baseline NoC.     

Lightweight fault localization combined with fault context to improve fault absolute rank

Network-on-Chip (NoC) is a promising replacement of bus architecture due to its better scalability. In state-of-the-art NoCs, each packet contains several fixed-length flits, which facilitates allocations of network resources but brings in many unused bits. In this paper, we propose a novel technique called Stealth-ACK to effectively address the above problem. Stealth-ACK leverages unused bits in head flits of non-ACK packets to carry and stealthily transmit ACK information. Such stealth transmissions of ACK information effectively reduce not only the amount of dedicated ACK packets on NoC, but also the number of unused bits in head flits of non-ACK packets, which significantly reduces wastes on NoC bandwidth. Experimental results show that Stealth-ACK averagely increases the throughput of 16 × 16 2-D mesh NoC by 11.9%, and averagely reduces the NoC latency by 34.8% on application traces of SPLASH-2. Moreover, Stealth-ACK only requires trivial hardware modification to basic router architectures, which incurs negligible power consumption and area cost.

     

An efficient numerical-based crosstalk avoidance codec design for NoCs

With technology scaling, crosstalk fault has become a serious problem in reliable data transfer through Network on Chip (NoC) channels. The effects of crosstalk fault depend on transition patterns appearing on the wires of NoC channels. Among these patterns, Triplet Opposite Direction (TOD) imposes the worst crosstalk effects. Crosstalk Avoidance Codes (CACs) are the overhead-efficient mechanisms to tackle TODs. The main problem of CACs is their high imposed overheads to NoC routers. To solve this problem, this paper proposes an overhead-efficient coding mechanism called Penultimate-Subtracted Fibonacci (PS-Fibo) to alleviate crosstalk faults in NoC wires. PS-Fibo coding mechanism benefits the novel numerical system that not only completely removes TODs but also, is applicable to a wide range of NoC channel widths. The PS-Fibo coding mechanism is evaluated using BookSim-2 and VHDL-based simulations in the terms of codec efficiency on the crosstalk fault reduction, codec power consumption, codec area occupation and network performance. Evaluation results, carried out for a wide range of NoC channel widths indicate that PS-Fibo can improve power consumption and area occupations of codec and NoC performance with respect to the other state-of-the-art coding mechanisms.