1-5hit |
Ryuta KAWANO Ryota YASUDO Hiroki MATSUTANI Michihiro KOIBUCHI Hideharu AMANO
Recently proposed irregular networks can reduce the latency for both on-chip and off-chip systems with a large number of computing nodes and thus can improve the performance of parallel applications. However, these networks usually suffer from deadlocks in routing packets when using a naive minimal path routing algorithm. To solve this problem, we focus attention on a lately proposed theory that generalizes the turn model to maintain the network performance with deadlock-freedom. The theorems remain a challenge of applying themselves to arbitrary topologies including fully irregular networks. In this paper, we advance the theorems to completely general ones. Moreover, we provide a feasible implementation of a deadlock-free routing method based on our advanced theorem. Experimental results show that the routing method based on our proposed theorem can improve the network throughput by up to 138 % compared to a conventional deterministic minimal routing method. Moreover, when utilized as the escape path in Duato's protocol, it can improve the throughput by up to 26.3 % compared with the conventional up*/down* routing.
Yali LI Hongma LIU Shengjin WANG
A brain-computer interface (BCI) translates the brain activity into commands to control external devices. P300 speller based character recognition is an important kind of application system in BCI. In this paper, we propose a framework to integrate channel correlation analysis into P300 detection. This work is distinguished by two key contributions. First, a coefficient matrix is introduced and constructed for multiple channels with the elements indicating channel correlations. Agglomerative clustering is applied to group correlated channels. Second, the statistics of central tendency are used to fuse the information of correlated channels and generate virtual channels. The generated virtual channels can extend the EEG signals and lift up the signal-to-noise ratio. The correlated features from virtual channels are combined with original signals for classification and the outputs of discriminative classifier are used to determine the characters for spelling. Experimental results prove the effectiveness and efficiency of the channel correlation analysis based framework. Compared with the state-of-the-art, the recognition rate was increased by both 6% with 5 and 10 epochs by the proposed framework.
Nen-Chung WANG Tzung-Shi CHEN Chih-Ping CHU
In this paper, we propose an efficient dual-tree-based multicasting scheme with three destination-switch partition strategies on irregular switch-based networks. The dual-tree-based routing scheme supports adaptive, distributed, and deadlock-free multicast on irregular networks with double channels. We first describe a dual-tree structure constructed from the irregular networks and prove that the multicasting based on such a structure is deadlock-free. Then, an efficient multicast routing algorithm with three destination-switch partition strategies: source-switch-based partition, destination-switch-based partition, and all-switches-based partition, is proposed. Finally, we perform simulations to evaluate our proposed algorithm under various impact parameters: system size, message length, and startup time. The experimental results show that the improved tree-based multicasting scheme outperforms the usual tree-based multicasting scheme. The dual-tree-based multicasting scheme with destination-switch-based partition is shown to be the best for all situations.
Deogkyoo LEE Daekeun MOON Ilgu YUN Hagbae KIM
Since components faults occurring at arbitrary places (primarily on the links) affect seriously network performance and reliability, the multicomputers operating in harsh environments should be designed to guarantee normal network-missions in presence of those faults. One solution to the end is a fault-tolerant routing scheme, which enables messages to safely reach their destinations avoiding failed links when transmission of messages is blocked by certain faults. In the paper, we develop a fault-tolerant routing algorithm with deadlock freedom in an n-dimensional meshed network, and validate its efficiency and effectiveness through proper simulations. The aspects of fault-tolerance is adopted by appending partial-adaptiveness and detouring to the e-cube algorithm, while using a wormhole routing for the backbone routing method. The phenomenon of deadlock incurred due to its adaptiveness is eliminated by classifying a physical channel into a couple of virtual channels.
Jinsoo KIM Ji-Yun KIM Hyunsoo YOON Seung Ryoul MAENG Jung Wan CHO
We propose a fault-tolerant routing algorithm for 2D meshes. Our routing algorithm can tolerate any number of concave fault regions. It is based on xy-routing and uses the concept of the fault ring/chain composed of fault-free elements surrounding faults. Three virtual channels per physical link are used for deadlock-free routing on a fault ring. Four virtual channels are needed for a fault chain. For a concave fault ring, fault-free nodes in the concave region have been deactivated to avoid deadlock in the previous algorithms, which results in excessive loss of the computational power. Our algorithm ensures deadlock-freedom by restricting the virtual channel usage in the concave region, and it minimizes the loss of the computational power. We also extend the proposed routing scheme for adaptive fault-tolerant routing. The adaptive version requires the same number of virtual channels as the deterministic one.