A Study of Leveraging Memory Level Parallelism for DRAM System on Multi-Core Architecture

Yuxuan Wang,Yingping Zhang,Xiaotian Zhang,Jian Yin,Licheng Chen 보안공학연구지원센터 2016 International Journal of Hybrid Information Techno Vol.9 No.4

DRAM system has been more and more critical on modern multi-core architecture where the Moore’s law has been made effect on increasing the number of cores integrated in a processor chip. The performance of DRAM system is usually measured in term of bandwidth and latency, which are regarded as inherently depending on Row Buffer Hit Rate (RBHR) according to previous studies. In this paper, we find that Memory Level Parallelism (MLP) exhibits a stronger correlation with the performance of DRAM system on multi-core/many-core architecture than RBHR, and promoting MLP significantly improves DRAM system performance. In order to exploit the MLP, we have evaluated various approaches including multi-bank, multi-row-buffers, multi-memory-controllers and the obsolete Virtual Channel Memory (VCM). The experimental results show that VCM is a better alternative to traditional DRAM chip on multicore/many-core architecture than the other three approaches because VCM has almost all the advantages of the others: 1) it can improve homogeneous workloads’ IPC by 2.21X on a 16-core system with 32 virtual channels due to leveraging unexploited MLP. 2) It can also promote Quality-of-Service (QoS) of DRAM system by removing unfairness while memory controllers serve memory requests. 3) It can save energy and has low area costs. Unfortunately, VCM, which was proposed in the late 1990s, faded away before multi-core/manycore became dominated. Therefore, we suggest memory chip vendors reconsider the VCM technology for multi-core architecture.

Combination of High-Density Lipoprotein Cholesterol and Lipoprotein(a) as a Predictor of Collateral Circulation in Patients With Severe Unilateral Internal Carotid Artery Stenosis or Occlusion

Shuyin Ma,Meijuan Zhang,Huiyang Qu,Yuxuan Cheng,Shuang Du,Jiaxin Fan,Qingling Yao,Xiao Dong Zhang,Mengying Chen,Nan Zhang,Kaili Shi,Yizhou Huang,Shuqin Zhan 대한신경과학회 2022 Journal of Clinical Neurology Vol.18 No.1

Background and Purpose Collateral circulation is considered an important factor affecting the risk of stroke, but the factors that affect collateral circulation remain unclear. This study was performed to identify the factors associated with collateral circulation, especially blood lipids. Methods The study involved patients who had undergone digital subtraction angiography and were confirmed as having severe unilateral stenosis or occlusion of the internal carotid artery (ICA). We classified the collateral circulation status of each patient as good (Grade 3 or 4) or poor (Grade 0, 1, or 2) according to the grading system of the American Society of Interventional and Therapeutic Neuroradiology/American Society of Interventional Radiology. We collected data on patients’ characteristics and identified the factors that affect collateral circulation. Results This study included 212 patients. The multivariate logistic regression analysis showed that the high-density lipoprotein cholesterol (HDL-C) concentration and a complete anterior half of the circle of Willis were independent protective factors for good collateral circulation, whereas elevated lipoprotein(a) [Lp(a)] and serum creatinine concentrations were independent risk factors for good collateral circulation. The area under the receiver operating characteristics curve (AUC) was 0.68 (95% confidence interval [CI], 0.61–0.76) for HDL-C and 0.69 (95% CI, 0.62–0.76) for Lp(a). A binary logistic regression model analysis of the joint factor of HDL-C and Lp(a) yielded an AUC of 0.77 (95% CI, 0.71–0.84). Conclusions In patients with severe unilateral ICA stenosis or occlusion, the combination of HDL-C and Lp(a) is a useful predictor of collateral circulation.

Selective absorption of H2S and CO2 from simulated coke oven gas by aqueous blends of N-methyldiethanolamine and tetramethylammonium glycine

Pan Zhang,Yuetong Zhao,Xiangfeng Tian,Yanxi Ji,Yuxuan Shu,Kun Fu,Dong Fu,Lemeng Wang 한국화학공학회 2022 Korean Journal of Chemical Engineering Vol.39 No.11

Tetramethylammonium glycine ([N1111][Gly]) can be completely ionized into cation [N1111]+ and anion [Gly] in aqueous solution. The anion contains an amino -NH2 and a carboxyl -COO, both of which can react with hydrogen sulfide (H2S). Therefore, [N1111][Gly] was used to promote the selective absorption of H2S in coke oven gas (COG) by N-methyldiethanolamine (MDEA). The absorption performance and selectivity of H2S in the aqueous solution of MDEA-[N1111][Gly] were investigated. The effects of MDEA mass fraction, [N1111][Gly] mass fraction, temperature, H2S partial pressure and CO2 partial pressure on the absorption capacity and selectivity were clarified. The results showed that an aqueous solution of MDEA-[N1111][Gly] has good selectivity for H2S in COG. The absorption capacity was large and the mass fraction of the solute in the absorbent reached more than 0.55, thereby having outstanding advantages in the aspects of saving energy consumption and operating cost and having a good application potential.

Hierarchically porous carbon derived from tobacco waste by one-step molten salt carbonization for supercapacitor

Liu Yuxuan,Cheng Xinhua,Zhang Shenghui 한국탄소학회 2022 Carbon Letters Vol.32 No.1

High-performance carbon materials were prepared via a one-step molten salt carbonization of tobacco waste used as electrode materials for supercapacitors. Carbon material prepared by carbonization for 3 h in molten CaCl2 at 850 °C exhibits hierarchically porous structure and ideal capacitive behavior. In a three-electrode configuration with 1 mol L−1 H2SO4 aqueous solution, it delivers specific capacitance of 196.5 F g−1 at 0.2 A g−1, energy density of 27.2 Wh kg−1 at 0.2 A g−1, power density of 983.5 W kg−1 at 2 A g−1, and excellent cyclic stability with 94% capacitance retention after 5000 charge–discharge cycles at 1 A g−1. Moreover, in a symmetrical two-electrode configuration with 6 mol L−1 KOH aqueous solution, it delivers specific capacitance of 111.1 F g−1 at 0.2 A g−1, energy density of 3.8 Wh kg−1 at 0.2 A g−1, and power density of 482.0 W kg−1 at 2 A g−1. The relationship between hierarchically porous structure and capacitive performance is also discussed.

Multiresponse Surfaces Optimization Based on Evidential Reasoning Theory

He, Zhen,Zhang, Yuxuan The Korean Society for Quality Management 2004 The Asian Journal on Quality Vol.5 No.1

During process design or process optimization, it is quite common for experimenters to find optimum operating conditions for several responses simultaneously. The traditional multiresponse surfaces optimization methods do not consider the uncertain relationship among these responses sufficiently. For this reason, the authors propose an optimization method based on evidential reasoning theory by Dempster and Shafer. By maximizing the basic probability assignment function, which indicates the degree of belief that certain operating condition is the solution of this multiresponse surfaces optimization problem, the desirable operating condition can be found.

Simulation of Permeation of Saturated Cement Paste Based on a New Meso-scale Pore Network Model

Yong Zhou,Yuxuan Yang,Bigya Gyawali,Weiping Zhang 한국콘크리트학회 2021 International Journal of Concrete Structures and M Vol.15 No.5

This paper presents the simulation of the permeation of saturated cement paste based on a novel pore network model. First, a 2D hydration model of cement particles was developed by extending the work of Zheng et al. 2005 to provide the background for the network construction. Secondly, the establishment of the pore network model and simulation of permeation of saturated cement paste were carried out. The irregular pores between any two hydrated cement particles were linearized with clear distances as the diameters of pores. The straight tubular pores were interconnected with one another to form the network model. During this process, the weighted Voronoi diagram was employed to operate on the graphical expression of the hydrated cement particles. Water permeation in saturated cement paste was simulated to verify the pore network model. Finally, the factors including water-cement ratio, reaction temperature, reaction time and cement particle size that would influence water permeation were numerically investigated.

Insight Into the Crystallinity of Chinese Ancient Silk by Synchrotron Radiation-Based and Conventional X-ray Diffraction Methods

Decai Gong,Xiaoning Zhang,Yuxuan Gong 한국문화재보존과학회 2020 보존과학회지 Vol.36 No.1

This study investigates the crystallinity distribution of ancient silk. Owing to the inherent multi-hierarchical structure of silk protein and the complicated structural changes that occur due to various burial environments, it is challenging but worthwhile to study ancient silk ageing behavior, which is based on the fact that ageing begins with a single fiber and then spreads to a whole fabric. Crystallinity was one of the most effective indicators found to reveal the ageing status of silk. Therefore, a synchrotron radiation-based X-ray diffraction(SR-XRD) method was employed to study the crystallinity distribution of single fibers of ancient silk unearthed from seven archaeological sites in China from historical periods including the warring states, Han dynasty, Song dynasty, and Ming dynasty. In comparison, the conventional X-ray diffraction method, which uses large amounts of samples, was also performed to determine the integral crystallinity of ancient silk. Thermal stability experiments by thermogravimetry(TG) as well as morphology observations by scanning electron microscopy(SEM) and optical microscopy(OM) all confirmed the deterioration of ancient silk. Moreover, the ageing mechanism of ancient silk was proposed with the assistance of an artificial ageing study. The results confirmed the effectiveness of SR-XRD as an ageing indicator, revealing the crystallinity distribution. This research could provide motivation to determine the deterioration status of ancient silk, and would also aid in explaining the fragility of ancient silk due to ageing.

Optimized interfacial compatibility of carbon fiber and epoxy resin via controllable thickness and activated ingredients of polydopamine layer

Wu Dongliang,Xing Yuxuan,Zhang Denglu,Hao Zhenna,Dong Qi,Han Yongqin,Liu Lei,Wang Maoju,Zhang Ruliang 한국탄소학회 2024 Carbon Letters Vol.34 No.1

In this work, the trend in the performance of carbon fiber (CF) and its composite during self-polymerization of polydopamine (PDA) at carbon fiber surface was investigated by varying the self-polymerization time of dopamine in an aqueous solution. Research has shown that the PDA coating elevated the surface roughness and polarity of the inert fiber. The tensile strength of single carbon fiber was significantly improved, especially after 9 h of polydopamine self-polymerization, increasing by 18.64% compared with that of desized carbon fiber. Moreover, the interlaminar shear strength (ILSS) of CF-PDA9-based composites was 35.06% higher than that of desized CF-based composites. This research will provide a deep insight into the thickness and activated ingredients of dopamine oxidation and self-polymerization on interfacial compatibility of carbon fiber/epoxy resin composites.

Identification of Key microRNAs in Diabetes Mellitus Erectile Dysfunction Rats with Stem Cell Therapy by Bioinformatic Analysis of Deep Sequencing Data

Kang Jiaqi,Song Yuxuan,Zhang Zhexin,Wang Shangren,Lu Yi,Liu Xiaoqiang 대한남성과학회 2022 The World Journal of Men's Health Vol.40 No.4

Purpose: Diabetes mellitus erectile dysfunction (DMED) is a common resulting complication of diabetes. Studies have shown mesenchymal stem cell (MSC)-based therapy was beneficial in alleviating erectile function of DMED rats. While the pathogenesis of DMED and the mechanism MSCs actions are unclear. Materials and Methods: We constructed a rat model of DMED with or without intracavernous injection of MSCs, and performed microRNA (miRNA) sequencing of corpora cavernosa tissues. Results: We identified three overlapping differentially expressed miRNAs (rno-miR-1298, rno-miR-122-5p, and rnomiR- 6321) of the normal control group, DMED group, and DMED+MSCs group. We predicted 285 target genes of three miRNAs through RNAhybrid and miRanda database and constructed a miRNA-target gene network through Cytoscape. Next, we constructed protein-protein interaction networks through STRING database and identified the top 10 hub genes with highest connectivity scores. Five GO terms including cellular response to growth factor stimulus (GO:0071363), ossification (GO:0001503), response to steroid hormone (GO:0048545), angiogenesis (GO:0001525), positive regulation of apoptotic process (GO:0043065), and one Reactome pathway (Innate Immune System) were significantly enriched by 10 hub genes using the Metascape database. We selected the GSE2457 dataset to validate the expression of hub genes and found only the expression of B4galt1 was statistically different (p<0.001). B4galt1 was highly expressed in penile tissues of diabetic rats and would be negatively regulated by rno-miR-1298. Conclusions: Three key miRNAs were identified in DMED rats with stem cell therapy and the miR-1298/B4GalT1 axis might exert function in stem cell therapy for ED.

An Enhanced Video Anomaly Detection System for Smart City Management

Yuechun Wang,Shufei Zhu,Yuxuan Zhao,Jie Zhang,Ka Lok Man 한국ITS학회 2022 한국ITS학회 학술대회 Vol.2022 No.06

Since the popularity of city monitoring systems has increased, surveillance videos have become more common. However, due to the massive increment of social event needs after lifting the coronavirus pandemic lockdown, the pressure of handling anomalies on government departments is growing with each passing day. In addition, surveillance cameras cannot record every corner of our city. There are still blind areas such as indoor environments and remote regions. Therefore, this paper introduces and develops an comprehensive system for video anomaly detection in smart city management, which is based on mobile phones, to enlarge the coverage of the current surveillance system. This system can capture and transmit abnormal videos. In addition, a deep learning method (VGG16) is implemented in the system to do video anomaly detection.