RESUMO
Introduction: Great knowledge was gained about the computational substrate of the brain, but the way in which components and entities interact to perform information processing still remains a secret. Complex and large-scale network models have been developed to unveil processes at the ensemble level taking place over a large range of timescales. They challenge any kind of simulation platform, so that efficient implementations need to be developed that gain from focusing on a set of relevant models. With increasing network sizes imposed by these models, low latency inter-node communication becomes a critical aspect. This situation is even accentuated, if slow processes like learning should be covered, that require faster than real-time simulation. Methods: Therefore, this article presents two simulation frameworks, in which network-on-chip simulators are interfaced with the neuroscientific development environment NEST. This combination yields network traffic that is directly defined by the relevant neural network models and used to steer the network-on-chip simulations. As one of the outcomes, instructive statistics on network latencies are obtained. Since time stamps of different granularity are used by the simulators, a conversion is required that can be exploited to emulate an intended acceleration factor. Results: By application of the frameworks to scaled versions of the cortical microcircuit model-selected because of its unique properties as well as challenging demands-performance curves, latency, and traffic distributions could be determined. Discussion: The distinct characteristic of the second framework is its tree-based source-address driven multicast support, which, in connection with the torus topology, always led to the best results. Although currently biased by some inherent assumptions of the network-on-chip simulators, the results suit well to those of previous work dealing with node internals and suggesting accelerated simulations to be in reach.
RESUMO
Real-time heterogeneous parallel embedded digital signal processor (DSP) systems process multiple data streams in parallel in a stringent time interval. This type of system on chip (SoC) requires the network on chip (NoC) to establish multiple symbiotic parallel data transmission paths with ultra-low transmission latency in real time. Our early NoC research PCCNOC meets this need. The PCCNOC uses packet routing to establish and lock a transmission circuit, so that PCCNOC is perfectly suitable for ultra-low latency and high-bandwidth transmission of long data packets. However, a parallel multi-data stream DSP system also needs to transmit roughly the same number of short data packets for job configuration and job execution status reports. While transferring short data packets, the link establishment routing delay of short data packets becomes relatively obvious. Our further research, thus, introduced PaCHNOC, a hybrid NoC in which long data packets are transmitted through a circuit established and locked by routing, and short data packets are attached to the routing packet and the transmission is completed during the routing process, thus avoiding the PCCNOC setup delay. Simulation shows that PaCHNOC performs well in supporting real-time heterogeneous parallel embedded DSP systems and achieves overall latency reduction 65% compared with related works. Finally, we used PaCHNOC in the baseband subsystem of a real 5G base station, which proved that our research is the best NoC for baseband subsystem of 5G base stations, which reduce 31% comprehensive latency in comparison to related works.
RESUMO
Sensor applications in internet of things (IoT) systems, coupled with artificial intelligence (AI) technology, are becoming an increasingly significant part of modern life. For low-latency AI computation in IoT systems, there is a growing preference for edge-based computing over cloud-based alternatives. The restricted coulomb energy neural network (RCE-NN) is a machine learning algorithm well-suited for implementation on edge devices due to its simple learning and recognition scheme. In addition, because the RCE-NN generates neurons as needed, it is easy to adjust the network structure and learn additional data. Therefore, the RCE-NN can provide edge-based real-time processing for various sensor applications. However, previous RCE-NN accelerators have limited scalability when the number of neurons increases. In this paper, we propose a network-on-chip (NoC)-based RCE-NN accelerator and present the results of implementation on a field-programmable gate array (FPGA). NoC is an effective solution for managing massive interconnections. The proposed RCE-NN accelerator utilizes a hierarchical-star (H-star) topology, which efficiently handles a large number of neurons, along with routers specifically designed for the RCE-NN. These approaches result in only a slight decrease in the maximum operating frequency as the number of neurons increases. Consequently, the maximum operating frequency of the proposed RCE-NN accelerator with 512 neurons increased by 126.1% compared to a previous RCE-NN accelerator. This enhancement was verified with two datasets for gas and sign language recognition, achieving accelerations of up to 54.8% in learning time and up to 45.7% in recognition time. The NoC scheme of the proposed RCE-NN accelerator is an appropriate solution to ensure the scalability of the neural network while providing high-performance on-chip learning and recognition.
RESUMO
Organizations managing high-performance computing systems face a multitude of challenges, including overarching concerns such as overall energy consumption, microprocessor clock frequency limitations, and the escalating costs associated with chip production. Evidently, processor speeds have plateaued over the last decade, persisting within the range of 2 GHz to 5 GHz. Scholars assert that brain-inspired computing holds substantial promise for mitigating these challenges. The spiking neural network (SNN) particularly stands out for its commendable power efficiency when juxtaposed with conventional design paradigms. Nevertheless, our scrutiny has brought to light several pivotal challenges impeding the seamless implementation of large-scale neural networks (NNs) on silicon. These challenges encompass the absence of automated tools, the need for multifaceted domain expertise, and the inadequacy of existing algorithms to efficiently partition and place extensive SNN computations onto hardware infrastructure. In this paper, we posit the development of an automated tool flow capable of transmuting any NN into an SNN. This undertaking involves the creation of a novel graph-partitioning algorithm designed to strategically place SNNs on a network-on-chip (NoC), thereby paving the way for future energy-efficient and high-performance computing paradigms. The presented methodology showcases its effectiveness by successfully transforming ANN architectures into SNNs with a marginal average error penalty of merely 2.65%. The proposed graph-partitioning algorithm enables a 14.22% decrease in inter-synaptic communication and an 87.58% reduction in intra-synaptic communication, on average, underscoring the effectiveness of the proposed algorithm in optimizing NN communication pathways. Compared to a baseline graph-partitioning algorithm, the proposed approach exhibits an average decrease of 79.74% in latency and a 14.67% reduction in energy consumption. Using existing NoC tools, the energy-latency product of SNN architectures is, on average, 82.71% lower than that of the baseline architectures.
RESUMO
Due to globalization in the semiconductor industry, malevolent modifications made in the hardware circuitry, known as hardware Trojans (HTs), have rendered the security of the chip very critical. Over the years, many methods have been proposed to detect and mitigate these HTs in general integrated circuits. However, insufficient effort has been made for hardware Trojans (HTs) in the network-on-chip. In this study, we implement a countermeasure to congeal the network-on-chip hardware design in order to prevent changes from being made to the network-on-chip design. We propose a collaborative method which uses flit integrity and dynamic flit permutation to eliminate the hardware Trojan inserted into the router of the NoC by a disloyal employee or a third-party vendor corporation. The proposed method increases the number of received packets by up to 10% more compared to existing techniques, which contain HTs in the destination address of the flit. Compared to the runtime HT mitigation method, the proposed scheme also decreases the average latency for the hardware Trojan inserted in the flit's header, tail, and destination field up to 14.7%, 8%, and 3%, respectively.
RESUMO
Deep learning has a better output quality compared with traditional algorithms for video super-resolution (SR), but the network model needs large resources and has poor real-time performance. This paper focuses on solving the speed problem of SR; it achieves real-time SR by the collaborative design of a deep learning video SR algorithm and GPU parallel acceleration. An algorithm combining deep learning networks with a lookup table (LUT) is proposed for the video SR, which ensures both the SR effect and ease of GPU parallel acceleration. The computational efficiency of the GPU network-on-chip algorithm is improved to ensure real-time performance by three major GPU optimization strategies: storage access optimization, conditional branching function optimization, and threading optimization. Finally, the network-on-chip was implemented on a RTX 3090 GPU, and the validity of the algorithm was demonstrated through ablation experiments. In addition, SR performance is compared with existing classical algorithms based on standard datasets. The new algorithm was found to be more efficient than the SR-LUT algorithm. The average PSNR was 0.61 dB higher than the SR-LUT-V algorithm and 0.24 dB higher than the SR-LUT-S algorithm. At the same time, the speed of real video SR was tested. For a real video with a resolution of 540×540, the proposed GPU network-on-chip achieved a speed of 42 FPS. The new method is 9.1 times faster than the original SR-LUT-S fast method, which was directly imported into the GPU for processing.
RESUMO
Hundreds of processor cores or modules are integrated into a single chip. The traditional bus or crossbar is challenged by bandwidth, scalability, and silicon area, and cannot meet the requirements of high end applications. Network-on-chip (NoC) has become a very promising interconnection structure because of its good scalability, predictable interconnect length and delay, high bandwidth, and reusability. However, the most available packet routing NoC may not be the perfect solution for high-end heterogeneous multi-core real-time systems-on-chip (SoC) because of the excessive latency and cache cost overhead. Moreover, circuit switching is limited by the scale, connectivity flexibility, and excessive overhead of fully connected systems. To solve the above problems and to meet the need for low latency, high throughput, and flexibility, this paper proposes PCCNoC (Packet Connected Circuit NoC), a low-latency and low-overhead NoC based on both packet switching (setting-up circuit) and circuit switching (data transmission on circuit), which offers flexible routing and zero overhead of data transmission latency, making it suitable for high-end heterogeneous multi-core real-time SoC at various system scales. Compared with typically available packet switched NoC, our PCCoC sees 242% improved performance and 97% latency reduction while keeping the silicon cost relatively low.
RESUMO
Traffic splitting enabled by Globally Asynchronous Locally Synchronous (GALS) Network-on-chip (NoC) brings multipath routing capability, which significantly increases link bandwidth at the cost of out-of-order packet delivery. Solving the packet reordering problem is one of the keys to ensure the quality of service (QoS) for NoC. However, traditional packet reordering approaches rely on local reorder buffer, causing on-chip hotspots, which aggravates chip aging and even leads to interconnection failures. In this paper, we present a multistage packet reordering (MPR) approach, which cannot only reduce the transmission latency but also effectively reduces hotspots caused by local reordering. Specifically, we propose multistage reordering buffer (MRB) by reusing channel buffers for implementing MPR. Experimental results show that our proposed approach achieved improved thermal efficiency with reduced hardware resource consumption.
RESUMO
Network on chip (NoC) is the main solution to the communication bandwidth of a multi-processor system on chip (MPSoC). NoC also brings more route requirements and is highly prone to errors caused by crosstalk. Crosstalk has become a major design problem in deep-submicron NoC communication design. Hence, a crosstalk error model and corresponding reliable system with error correction code (ECC) are required to make NoC communication reliable. In this paper, a reliability system evaluation model (RSE) of NoC communication with analysis from backend to frontend has been proposed. In the backend, a crosstalk error rate model (CER) is established with a three-wire RLC coupling model and timing constraints. The CER is used to establish functional relations between interconnect spacing, length and signal frequency, and test system reliability. In the frontend, a reliability system performance model (RSP) is established with a CER, reliability method cost and bandwidth. The RSE summarizes the frontend and backend model. In order to verify the RSE model, we propose a reliability system with a hybrid automatic repeat request technique (RSHARQ). Simulation demonstrates that the CER model is close to real circuit design. Through the CER and RSP model, the performance of RSHARQ could be simulated.
RESUMO
This article considers the usage of circulant topologies as a promising deadlock-free topology for networks-on-chip (NoCs). A new high-level model, Newxim, for the exploration of NoCs with any topology is presented. Two methods for solving the problem of cyclic dependencies in circulant topologies, which limit their applications for NoCs due to the increased possibility of deadlocks, are proposed. The first method of dealing with deadlocks is universal and applicable to any topology; it is based on the idea of bypassing blocked sections of the network on an acyclic subnetwork. The second method-Ring-Split-takes into account the features of circulant topologies. The results of high-level modeling and comparison of the peak throughput of NoCs for circulant and mesh topologies using deadlock-free routing algorithms are presented. It was shown that a new approach for routing in circulants (compared to mesh topology) shows up to 59% better network throughput with a uniform distribution of network load.
RESUMO
Networks-on-Chip (NoCs) have become the de-facto on-chip interconnect for multi/manycore systems. A typical NoC router is made up of buffers used to store packets that are unable to advance to their desired destination. However, buffers consume significant power/area and are often underutilized, especially in cases of applications with non-uniform traffic patterns thus leading to performance degradation for such applications. To improve network performance, the Roundabout NoC (R-NoC) concept is considered. R-NoC is inspired by real-life multi-lane traffic roundabouts and consists of lanes that are shared by multiple input/output ports to maximize buffering resource utilization. R-NoC relies on router-internal adaptive routing that decides the lane path based on back pressure. Back pressure makes it possible to assess lane utilization and route packets accordingly. This is made possible thanks to the use of elastic buffers for control flow, a clever type of handshaking in a way similar to asynchronous circuits. Another prominent feature of R-NoC is that internal routing and arbitration are completely distributed which allows for significant freedom in deciding internal router topology and parameters. This work leverages this property and proposes novel yet unexplored configurations for which an in-depth evaluation of corresponding implementations on 45 nm CMOS technology is given. Each configuration is evaluated performance and power-wise on both synthetic and real application traffic. Several R-NoC configurations are identified and demonstrated to provide very significant performance improvements over standard mesh configurations and a typical input-buffered router, without compromising area and power consumption. Exploiting the distributed nature of R-NoC routers, a diagonally-linked configuration is then proposed which incurs moderate area overhead and features yet better performance and energy efficiency.
RESUMO
The die-stacking structure of 3D network-on-chips (3D NoC) leads to high power density and unequal thermal conductance between different layers, which results in low reliability and performance degradation of 3D NoCs. Congestion-aware adaptive routing, which is capable of balancing the network's traffic load, can alleviate congestion and thermal problems so as to improve the performance of the network. In this study, we propose a traffic- and thermal-aware Q-routing algorithm (TTQR) based on Q-learning, a reinforcement learning method. The proposed algorithm saves the local traffic status and the global temperature information to the Q1-table and Q2-table, respectively. The values of two tables are updated by the packet header and saved in a small size, which saves the hardware overhead. Based on the ratio of the Q1-value to the Q2-value corresponding to each direction, the packet's output port is selected. As a result, packets are transferred to the chosen path to alleviate thermal problems and achieve more balanced inter-layer traffic. Through the Access Noxim simulation platform, we compare the proposed routing algorithm with the TAAR routing algorithm. According to experimental results using synthetic traffic patterns, our proposed methods outperform the TAAR routing algorithm by an average of 63.6% and 41.4% in average latency and throughput, respectively.
RESUMO
A self-reconfigurable Network-on-Chip (NoC) architecture that supports anticipative Quality of Service (QoS) control with penetrative switch ability is proposed to enhance the performance of bidirectional-channel NoC communication while supporting prioritized packet transmission services. The anticipative QoS control not only allows each communication channel to be dynamically self-configured to transmit flits in either direction for a better channel utilization of on-chip hardware resources, but also enhances the latency performance for QoS services. The proposed anticipative control is based on penetratingly observing channel direction requests of routers that is two hops away from the current one. The added ability enables a router to allocate high-priority packets to a dedicated virtual channel and then rapidly bypass it to the next destination router. The provided flexibility of packet switch promises better channel bandwidth utilization, lower packet delivery latency, and furthermore guarantees the high-priority packets being served with a better QoS. Accordingly, in this paper, an enhanced NoC architecture supporting the hybrid anticipative QoS, penetrative switch, and bidirectional-channel control, namely Anticipative QoS Bidirectional-channel NoC (AQ-BiNoC) is presented. Tested with cycle-accurate synthetic traffic patterns, significant performance enhancement has been observed when the proposed AQ-BiNoC architecture is compared against conventional NoC designs.
RESUMO
This paper introduces the concept of smart radio environments, currently intensely studied for wireless communication in metasurface-programmable meter-scaled environments (e.g., inside rooms), on the chip scale. Wireless networks-on-chips (WNoCs) are a candidate technology to improve inter-core communication on chips but current proposals are plagued by a dilemma: either the received signal is weak, or it is significantly reverberated such that the on-off-keying modulation speed must be throttled. Here, this vexing problem is overcome by endowing the wireless on-chip environment with in situ programmability which enables the shaping of the channel impulse response (CIR); thereby, a pulse-like CIR shape can be imposed despite strong multipath propagation and without entailing a reduced received signal strength. First, a programmable metasurface suitable for integration in the on-chip environment ("on-chip reconfigurable intelligent surface") is designed and characterized. Second, its configuration is optimized to equalize selected wireless on-chip channels "over the air." Third, by conducting a rigorous communication analysis, the feasibility of significantly higher modulation speeds with shaped CIRs is evidenced. The results introduce a programmability paradigm to WNoCs which boosts their competitiveness as complementary on-chip interconnect solution.
Assuntos
Tecnologia sem Fio , Redes de Comunicação de Computadores , RádioRESUMO
Real-time image processing and computer vision systems are now in the mainstream of technologies enabling applications for cyber-physical systems, Internet of Things, augmented reality, and Industry 4.0. These applications bring the need for Smart Cameras for local real-time processing of images and videos. However, the massive amount of data to be processed within short deadlines cannot be handled by most commercial cameras. In this work, we show the design and implementation of a manycore vision processor architecture to be used in Smart Cameras. With massive parallelism exploration and application-specific characteristics, our architecture is composed of distributed processing elements and memories connected through a Network-on-Chip. The architecture was implemented as an FPGA overlay, focusing on optimized hardware utilization. The parameterized architecture was characterized by its hardware occupation, maximum operating frequency, and processing frame rate. Different configurations ranging from one to eighty-one processing elements were implemented and compared to several works from the literature. Using a System-on-Chip composed of an FPGA integrated into a general-purpose processor, we showcase the flexibility and efficiency of the hardware/software architecture. The results show that the proposed architecture successfully allies programmability and performance, being a suitable alternative for future Smart Cameras.
Assuntos
Inteligência Artificial , Processamento de Imagem Assistida por Computador , Computadores , SoftwareRESUMO
Network-on-Chip is a good approach to working on intra-chip communication. Networks with irregular topologies may be better suited for specific applications because of their architectural nature. A good design space exploration can help the design of the network to obtain more optimized topologies. This paper proposes a way of optimizing networks with irregular topologies through the use of a genetic algorithm. The network proposed here has heterogeneous routers that aim to optimize the network and support applications with real-time tasks. The goal is to find networks that are optimized for average latency and percentage of real-time packets delivered within the deadline. The results show that we have been able to find networks that can deliver all the real-time packets, obtain acceptable latency values, and shrink the chip area.
RESUMO
Mapping application task graphs on intellectual property (IP) cores into network-on-chip (NoC) is a non-deterministic polynomial-time hard problem. The evolution of network performance mainly depends on an effective and efficient mapping technique and the optimization of performance and cost metrics. These metrics mainly include power, reliability, area, thermal distribution and delay. A state-of-the-art mapping technique for NoC is introduced with the name of sailfish optimization algorithm (SFOA). The proposed algorithm minimizes the power dissipation of NoC via an empirical base applying a shared k-nearest neighbor clustering approach, and it gives quicker mapping over six considered standard benchmarks. The experimental results indicate that the proposed techniques outperform other existing nature-inspired metaheuristic approaches, especially in large application task graphs.
RESUMO
Network-on-Chips with simple topologies are widely used due to their scalability and high bandwidth. The transmission latency increases greatly with the number of on-chip nodes. A NoC, called single-cycle multi-hop asynchronous repeated traversal (SMART), is proposed to solve the problem by bypassing intermediate routers. However, the bypass setup request of SMART requires additional pipeline stages and wires. In this paper, we present a NoC with rapid bypass channels that integrates the bypass information into each flit. In the proposed NoC, all the bypass requests are delivered along with flits at the same time reducing the transmission latency. Besides, the bypass request is unicasted in our design instead of broadcasting in SMART leading to a great reduction in wire overhead. We evaluate the NoC in four synthetic traffic patterns. The result shows that the latency of our proposed NoC is 63.54% less than the 1-cycle NoC. Compared to SMART, more than 80% wire overhead and 27% latency are reduced.
RESUMO
Network-on-chip (NoC) architectures have become a popular communication platform for heterogeneous computing systems owing to their scalability and high performance. Aggressive technology scaling makes these architectures prone to both permanent and transient faults. This study focuses on the tolerance of a NoC router to permanent faults. A permanent fault in a NoC router severely impacts the performance of the entire network. Thus, it is necessary to incorporate component-level protection techniques in a router. In the proposed scheme, the input port utilizes a bypass path, virtual channel (VC) queuing, and VC closing strategies. Moreover, the routing computation stage utilizes spatial redundancy and double routing strategies, and the VC allocation stage utilizes spatial redundancy. The switch allocation stage utilizes run-time arbiter selection. The crossbar stage utilizes a triple bypass bus. The proposed router is highly fault-tolerant compared with the existing state-of-the-art fault-tolerant routers. The reliability of the proposed router is 7.98 times higher than that of the unprotected baseline router in terms of the mean-time-to-failure metric. The silicon protection factor metric is used to calculate the protection ability of the proposed router. Consequently, it is confirmed that the proposed router has a greater protection ability than the conventional fault-tolerant routers.
RESUMO
In this paper we propose and analyze various approaches to organizing routing in a triple loop circulant topologies as applied to networks-on-chip: static routing based on universal graph search algorithms, such as Dijkstra's algorithm and a possible implementation using Table routing; algorithms created analytically based on an engineering approach with taking into account the structural features of triple loop circulant graphs (Advanced clockwise, Direction selection); an algorithm created on the basis of a mathematical analysis of graph structure and solving the problem of enumerating coefficients at generators (Coefficients finding algorithm). Efficiency, maximum graph paths, occupied memory resources, and calculation time of the algorithms developed are estimated. Comparison of various variants of the algorithms is made and recommendations on their application for the development of networks-on-chip with triple loop circulant topologies are given. It is shown that Advanced clockwise and Direction selection algorithms guarantee that the packet reaches the destination node, but often in more steps than the shortest path. Nevertheless, they themselves are simpler and require less hardware resources than other algorithms. In turn, Coefficients finding algorithm has great computational complexity, but is optimal and, in comparison with Dijkstra's algorithm, is much simpler for RTL implementation which reduces network-on-chip routers resources cost.