Graph traversal is a widely used algorithm in a variety of fields, including social networks, business analytics, and high-performancecomputing and so on. Graph traversal on single nodes has been well studied and opt...
详细信息
Graph traversal is a widely used algorithm in a variety of fields, including social networks, business analytics, and high-performancecomputing and so on. Graph traversal on single nodes has been well studied and optimized on modern CPU architectures. Now, heterogeneous computing is becoming more and more popular and CPU+MIC is a typical heterogeneous architectures. The Intel MIC(Many Integrated Core) has up to 57 cores and hasn't been fully evaluated for graph traversal. When use a MIC to traverse a graph, the MIC may suffer from loading imbalance for the reason that the degree of vertexes in a graph may differs very much, which can degrade system performance. So in this paper, an algorithmic design and optimization techniques are presented to load balancing in MIC. About the optimization design, the main idea is that treat the vertexes with big degree and the vertexes with small degree separately. For this reason, some adjustments will be made to existing algorithms and data structures. It has achieved almost big performance improvements over the BFS algorithm without loading balancing in MIC as shown in section VI. We believe that this novel algorithm can be successfully applied to a broader class of graph algorithms with many MIC cores.
The phonon properties, electronic structures and optical properties of some novel carbon allotropes including monolayer penta-graphene (PG), double-layer PG and T12-carbon were investigated using the first-principles ...
详细信息
Unpredictable power outages in NAND flashbased Solid State Drives (SSDs) may cause system failure or reliability problems. Capacitors are widely adopted as the interim power supplier when power interruption happens. H...
详细信息
ISBN:
(纸本)9781479989386
Unpredictable power outages in NAND flashbased Solid State Drives (SSDs) may cause system failure or reliability problems. Capacitors are widely adopted as the interim power supplier when power interruption happens. However, since the energy provided by backup capacitors is limited, and the capacitance of a capacitor will gradually degrade with time, it is imperative to improve the efficiency and reliability of the backup process for SSDs with capacitors. This paper presents a novel backup scheme called SmartBackupfor MLC NAND flash-based SSDs with backup capacitors. In SmartBackup, by exploiting the feature of the discharge process of capacitors, all available SSD channels are fully utilized and dynamically adjusted based on the available voltage, so data can be efficiently stored in a reliable manner. Moreover, in SmartBackup, to write data in a time/energyefficient manner with better reliability, we adopt a fast and reliable programming strategy by only writing LSB (Least Significant Bit) pages in MLC NAND flash. To the best of our acknowledge, this is the first work to jointly utilize the property of the multi-page architecture of MLC NAND flash and the discharge characteristic of capacitors to improvethe efficiency and reliability of the backup process in SSDswith backup capacitors. The experimental results show thatSmartBackup can effectively accelerate the backup processand improve the reliability.
This paper describes a parallel discrete event simulator, Neuron Time Warp-Multi Thread (NTW-MT), developed for the simulation of reaction diffusion models of neurons. The simulator was developed as part of the NEURON...
详细信息
We consider networks where mobile sensor/actor nodes move to specific locations in order to conduct data collection or deliver a response to an event. The challenge is to find the best tour for the mobile nodes in ord...
详细信息
ISBN:
(纸本)9781479959532
We consider networks where mobile sensor/actor nodes move to specific locations in order to conduct data collection or deliver a response to an event. The challenge is to find the best tour for the mobile nodes in order to visit the given set of locations. In this paper, the objective of the optimization is to extend the node lifetime by emphasizing both path efficiency and balanced energy consumption when identifying and assigning tours to mobile nodes. Compared to existing schemes in the literature, we consider the initial position of mobile sensors when determining the tours. We formulate the optimization as a balanced multi-salesman travel problem and propose a solution based on a two-step approach. First, we determine the shortest tour that includes all event locations by forming the Hamiltonian cycle. Then, we formulate the optimal partitioning of such a cycle as a linear program (LP) where the objective is to reduce the tour length while minimizing the maximum tour a node has to make. For scalability and to expedite convergence, we propose a method for solving the LP formulation based on Branch & Price algorithm. The simulation results confirm the effectiveness of our optimization formulation and the advantage of our solution compared to competing schemes.
Electromagnetic interference (EMI) issues are becoming crucial for three-dimensional integrated components which combine multi-core Systems-On-Chip, multi-band radio frequency circuits (RF), Giga-bit memories, as well...
详细信息
Electromagnetic interference (EMI) issues are becoming crucial for three-dimensional integrated components which combine multi-core Systems-On-Chip, multi-band radio frequency circuits (RF), Giga-bit memories, as well as advanced analog circuits. In this first paper, the technology roadmap towards 3D-ICs is illustrated and the associated EMC challenges are described. This first paper also focuses on the role of interposers, associated electrical parasitics, and addresses concerns about signal and power integrity.
Most recently, gesture recognition has increasingly attracted intense academic and industrial interest due to its various applications in daily life, such as home automation, mobile games. Present approaches for gestu...
详细信息
Most recently, gesture recognition has increasingly attracted intense academic and industrial interest due to its various applications in daily life, such as home automation, mobile games. Present approaches for gesture recognition, mainly including vision-based, sensor-based and RF-based, all have certain limitations which hinder their practical use in some scenarios. For example, the vision-based approaches fail to work well in poor light conditions and the sensor-based ones require users to wear devices. To address these, we propose WiG in this paper, a device-free gesture recognition system based solely on Commercial Off-The-Shelf (COTS) WiFi infrastructures and devices. Compared with existing Radio Frequency (RF)-based systems, WiG stands out for its systematic simplicity, extremely low cost and high practicability. We implemented WiG in indoor environment and conducted experiments to evaluate its performance in two typical scenarios. The results demonstrate that WiG can achieve an average recognition accuracy of 92% in line-of-sight scenario and average accuracy of 88% in the none-line-of sight scenario.
In this paper, a unified model for time-dependent Maxwell equations in dispersive media is considered. The space-time DG method developed in [29] is applied to solve the un-derlying problem. Unconditional L2-stability...
详细信息
In this paper, a unified model for time-dependent Maxwell equations in dispersive media is considered. The space-time DG method developed in [29] is applied to solve the un-derlying problem. Unconditional L2-stability and error estimate of order O?τr+1+hk+1/2? are obtained when polynomials of degree at most r and k are used for the temporal dis-cretization and spatial discretization respectively. 2-D and 3-D numerical examples are given to validate the theoretical results. Moreover, numerical results show an ultra-convergence of order 2r+1 in temporal variable t.
This paper reviews the accuracy of the Simplified Bernoulli Trial (SBT) algorithm and its variants, i.e., SBT-TAS (SBT on transient adaptive subcells) and ISBT (intelligence SBT) in the simulation of a wide spectrum o...
This paper reviews the accuracy of the Simplified Bernoulli Trial (SBT) algorithm and its variants, i.e., SBT-TAS (SBT on transient adaptive subcells) and ISBT (intelligence SBT) in the simulation of a wide spectrum of rarefied flow problems, including collision frequency ratio evaluation in the equilibrium condition, comparison of the Sonine-polynomial coefficients prediction in the Fourier flow with the theoretical prediction of the Chapman-Enskog expansion, accurate wall heat flux solution for the Fourier flow in the early slip regime, and hypersonic flows over cylinder and biconic geometries. We summarize advantages and requirements that utilization of the SBT collision families brings to a typical DSMC solver.
暂无评论