Search

More engaging and accessible solutions that offer outstanding user experience are needed. The virtual desktop infrastructure (VDI) is the first user access part of the cloud. Therefore, the performance of the VDI has a significant influence on the network. The graphics processing unit (GPU) can accelerate performance by assigning each virtual machine a dedicated GPU for performance-boosting to improve the user experience. In this paper, we propose a high-performance VDI using an integration of a GPU on OpenStack with a PCI pass-through. It might have certain benefits over using virtualized hardware, such as lower latency, better performance, and more functionality. The evaluation and comparison of the computer processing unit (CPU) and GPU are presented among the different benchmark tools.

This study aimed to identify the customers’ requirements for a hand rehabilitation exoskeleton device using the Kansei Engineering approach and propose improvements. The Kansei Engineering approach was used to extract the design attributes for a hand rehabilitation exoskeleton device by conducting an in-depth interview with 3 physical therapists. The Kano questionnaire survey investigated 25 participants with hand rehabilitation experience. The Kano classification results found three attractive factors and corresponded it to the design attributes of the evaluation grid method results that were suitable (size), flexible (controllability), and light (weight). The principal component analysis results indicated three components can be considered as the improvement categories; functions, structure-texture, and appearance of the hand exoskeleton device. The new functional and structural designs were proposed in a previous study. This study collects 110 male participants’ hand anthropometric data to establish a hand sizing system as a design improvement (suitable attribute). The 3D and manual measurements were obtained for 63 hand dimensions with an open palm posture. The 14 dorsal length dimensions with three bent finger postures were also manually measured. The results showed that the $${\text{MAD}}_{\text{Precision}}$$ (mean absolute difference) of 3D measurement is smaller than that of manual measurement, indicating better reproducibility of 3D scanning. The $${\text{MAD}}_{\text{Accuracy}}$$ of all dimensions are within the 1 mm of the referenced ISO 20685 criteria. Significant differences were found in the 14 dorsal hand length, showing that databases should explicitly state the aspect (dorsal or palmar) where dimensions have been of measurement. The 77 hand dimensions were taken into consideration to develop a hand sizing system using the clustering results. The two-level self-organizing map method was performed and 3 handshape types and 8 sizes were extracted. The coverage rate of the sizing system was over 85%. These results provide useful information for designing and manufacturing for hand-related products.

The virtual desktop infrastructure (VDI) service for the render farm platform at National Center for High-performance Computing (NCHC) was built to provide users with remote access to the render farm. Users from the film industry can utilize this platform throughout production, giving them the benefit of working on any regular laptop but with a dedicated Graphics Processing Unit (GPU). This paper describes an attempt at using NCHC’s VDI service for Augmented Reality (AR) and Virtual Reality (VR) to avoid the expensive equipment required.

Military helmets are designed specifically to protect soldiers’ heads during combat. Poor helmet fit has been reported to decrease the helmet’s comfort and the ability to provide safety protection. To increase the safety and comfort of soldiers wearing military helmets, the head size and pad design should be considered. Thus, the aim of this study is to evaluate the proper thickness of pad design for fit and comfort of military helmets. A total of 20 male participants were recruited in this study. A digital caliper and three-dimensional (3D) head scanner were used to measure participants’ helmet standoff distance (SOD) with two pad thickness levels (10 and 20 mm) placed in the same helmet shell. A 3D head-scanning system was used to obtain the 3D head shape model and to measure the head dimensions. The CATIA software was used to calculate the SOD between the 3D head-scanning model and the military helmet to assess helmet fit. The SOD results showed that the contact pressures for the two pad thicknesses were significantly different in five regions. The pad with a thickness of 10-mm had less contact pressure. Thus, the pad with a thickness of 10-mm is recommended because it can provide better fit and comfort for a military helmet. The findings of this study provide useful references for designing and developing head-related products in the future.

The pressure swing adsorption (PSA) process was used to capture carbon dioxide (CO2) from the flue gas of a coal-fired power plant to reduce CO2 emissions. Herein, CO2 was captured from flue gas using the PSA process for at least 85 vol% CO2 purity and with the other exit stream from the process of more than 90 vol% N2 purity. The extended Langmuir–Freundlich isotherm was used for calculating the equilibrium adsorption capacity, and the linear driving force model was used to describe the gas adsorption kinetics. We compared the results of breakthrough curves obtained through experiments and simulations to verify the accuracy of the mass transfer coefficient. The flue gas obtained after desulphurization and water removal (13.5 vol% CO2 and 86.5 vol% N2) from a subcritical 1-kW coal-fired power plant served as the feed for the designed three-bed, nine-step PSA process. To determine optimal operating conditions for the process, the central composite design (CCD) was used. After CCD analysis, optimal operating conditions with a feed pressure of 3.66 atm and a vacuum pressure of 0.05 atm were obtained to produce a bottom product with a CO2 purity of 89.20 vol% and a recovery of 88.20%, and a top product with a N2 purity of 98.49 vol% and a recovery of 93.56%. The mechanical energy consumption was estimated to be 1.17 GJ/t-CO2.

Existing nonvolatile ternary content-addressable-memory (nvTCAM) suffers from limited word-length (WDL), large write-energy ( $E_{W})$ and search-energy ( $E_{S})$ , and large cell area (A). This paper develops a 3T1R nvTCAM cell using a single multiple-level cell (MLC)-resistive RAM (ReRAM) device to achieve long WDL, lower $E_{W}$ and $E_{S}$ , and reduced cell area. Two peripheral control schemes were developed, dual-replica-row self-timed and invalid-entry power consumption suppression (IEPCS), for the suppression of dc current in 3T1R nvTCAM cells in order to reduce $E_{S}$ . Two versions of the IEPCS scheme were developed (basic and charge-recycle-controlled) to alter the tradeoff between area overhead and power consumption in the updating of invalid-bits. A $128~\text {b} \times 64$ b 3T1R nvTCAM macro was fabricated using back-end-of-line ReRAM under 90-nm CMOS process. The fabricated MLC-based 3T1R nvTCAM macro achieved sub-1-ns search-delay and sub-6-ns wake-up time with supply voltage of 1 V and WDL = 64 b.

This paper presents an experimental cloud service framework for design evaluation of personalized footwear in augmented reality (AR) via networks. The service allows users to ubiquitously perceive themselves trying on three-dimensional (3D) shoe models in a video stream. They upload a clip of feet motion recorded by a commercial depth camera to the cloud. A new clip is generated to display the try-on process and made available to specified receivers via video streaming on a mobile device. The framework design emphasizes making most use of open-source software and off-the-shelf technologies commercially available. A prototyping cloud system implementing the framework demonstrates the practical value of virtual footwear try-on as AR as a service (ARaaS). This experimental study realizes the idea of human-centric design evaluation in modern e-commerce. The cloud framework may provide a feasible example to improve the usability for real-time applications of AR.

In recent years, Information industry trends gradually towards the clouds. Various clouds are related technologies and applications are growing with the popularity of cloud technology. While cloud virtual desktop is the first user access part of the cloud and the performance of cloud virtual desk of the network and GPU has been implicated. A smooth cloud virtual desktop is a very important issue in this environment. In this study, we proposed an integration of GPU on OpenStack and through PCI Pass-Through way to solve the problem of virtual machine display performance by allocating GPU resources for virtual machine. For the experimental, we will use different benchmarking software to measure the virtual machine. In the experiment, adjust the different number of vCPUs to measure the different cases performance of the FPS virtual machine (Frame per Second). Through this experimentation, we found out there are three different processing drivers for the virtual desktop.

This study aimed to establish a head anthropometric database for the Taiwanese population and to compare two clustering analysis results for separating head-shape clusters as well as generating sizing systems. Two-stage clustering and two-level self-organizing map (SOM) methods were both applied. The PCA results indicated four components that can explain over 85% of the variability. The two-stage clustering grouped the head anthropometric data into four head shape types, in which twelve sizes were extracted. The two-level SOM method identified three head shape types, in which nine sizes were extracted. The coverage of both methods reached over 80%. Comparing the results, the two-level SOM method is superior to the two-stage clustering method. The findings of this study demonstrated that the two-level SOM method is an effective analysis tool with the ability to visualize latent spatial structures and characteristics within large datasets. These results provide useful information for designing and manufacturing head-related products.

Memristor devices are promising as high-density computing logic for non-high-speed normally off applications using heterogeneous technologies. This brief proposes a set-triggered-parallel-reset memristor logic (STPR-ML)-based on back-end-of-line-based bipolar-type memristor devices to reduce the area overhead and increase compatibility with standard and nonstandard CMOS processes. The proposed STPR-ML has two subsets: with and without nonvolatile storage capability. Two types of physical structures are also proposed for STPR-ML: 2-D and 3-D. A test chip using 2-D $\hbox{HfO}_{x}$ memristor devices was fabricated using a CMOS process to confirm the functionality of the nand , nor , and xor gates used in STPR-ML. The proposed STPR-ML achieves a $2\sim 8\times$ reduction in area compared with CMOS logic gates with four or more inputs.

This work focuses on how to provide the best virtual desktop technology, virtualization performance and hardware in the computer classroom. There are using the three popular virtualization technologies in the same environment to establish the same conditions, pure performance comparison: VMware Horizon, Citrix XenDesktop, and Microsoft VDI. Test software used to the current popular software, system performance including a test CPU and RAM PassMark Performance Test,and observe the vCenter Server performance changes and record statistics, to determine the advantages and disadvantages of various virtualization technologies. Finally, hardware support exclusive GPU virtualization product - Nvidia GRID K2 through GPU virtualization technology sharing and vGPU to carry on VMware vSphere, via Cinebench, YouTube 1080p high-quality video for bandwidth flow test, and use 3DMark Fire Strike Extreme to enhance the performance of virtual desktop, and bring users to enjoy higher-quality cloud services. The results of this study show that virtual desktop technologies are good in different field. On the other hand, the hardware performance breakthrough Moore’s Law of this era, the single CPU has more number of cores, and the memory capacity of the larger single,and provided by the GPU for graphics acceleration processing capabilities, the performance of server has been perfect than the previous. Therefore, the quality of cloud services is also greatly improved. The study shows that the virtual test can be imported into the classroom through the stress test in the experiment. To make sure the its performance and cost are really suitable for classroom imported desktop virtualization system.

In this paper, we propose a new secret sharing method for Chinese characters based on non-negative matrix factorization (NMF). Chinese characters can be regarded as binary images, but traditional NMF is only suitable for gray-level images. Therefore, this paper proposes a modified version of NMF for dealing with Chinese characters in binary images. The modified NMF decomposes a Chinese character into several parts, making each part unreadable to ensure that secrecy can be preserved. Experimental results show the decomposed results and the feasibility of the proposed method.

Computer-generated imagery (CGI) is becoming integral to a movie's story and appeal, and even dominates the film's success at box office. Currently the CGI realism is evaluated by post-production supervisors, and few objective realism assessments focus on this area. This paper investigates enhanced feature learning and classifier training for CGI assessment by deep learning. A training-set-selection method is proposed to select proper samples, which is crucial to deep learning training. Then the selected samples are converted into entropy images with enhanced features. We adopt a convolutional neural network for feature learning and classifier training to estimate the realism of CGI. Experimental results show that the developed matric has acceptable accuracy when compared to the grout truth. In addition, the rating result of the proposed assessment is very close to that of human visual perception.

We propose a novel exoskeleton for grasping hand rehabilitation based on anthropometry. The proposed design has one degree of freedom (DOF) for each finger, yielding coordinated movement across the distal interphalangeal (DIP), proximal interphalangeal (PIP), and metacarpophalangeal (MCP) joints for each finger. The dimension of each segment is determined by hand anthropometric data obtained from measurements. Each finger is controlled by one motor to allow for independent movement of each finger, which is fundamental for hand dexterity. The design was guided by a proposed mechanical model (the exo-finger model) which was verified by simulation and validated by the movement recorded by prototype fingers. It is concluded that, in the present study, anthropometry-based structural design provides a framework for the development of exoskeletal robotic devices.

Many search engines or filters for the internet-of-things and big-data employ ternary content-addressable-memory (TCAM) to suppress power consumption in the transmission of data between end-devices and servers. Nonvolatile TCAMs (nvTCAM) are designed to achieve zero standby power with smaller area overhead and faster power off/on operations than those found in conventional TCAM+NVM 2-macro schemes. In this paper, we discuss the challenges involved in the design of nvTCAMs and propose a universal 5T2R nvTCAM cell with tolerance for the various R-ratios and write parameters associated with emerging memory devices. A 128×64b-nvTCAM macro was fabricated using HfO ReRAM and a 90nm-CMOS process for concept verification.

Based on the high penetration rate of mobile devices, reducing power consumption for mobile devices have become increasingly important. In current technology, statistics show that the display of the mobile device dominates the major power consumption of the device (e.g., up to 40% of the total power). Therefore, to efficiently reduce the display power consumption while maintaining the image quality, this paper proposes a novel contrast-aware backlight control algorithm for mobile devices. By transforming the color space for a displayed image from RGB (red, green, blue) to HSI (hue, saturation, intensity), we first calculate the RMS (root mean square) contrast for the image. Then we estimate the "brightness factor" for adjusting the image based on the RMS contrast and the employed HVS (human vision system)-based visual quality assessment technique. At the same time, the estimated brightness factor is used to determine the screen backlight brightness value of the device. As a result, the best tradeoff between the image display quality and the power consumption can be achieved. Our experimental results have demonstrated that 10∼40% reduction of backlight brightness can be achieved, resulting in power consumption of 8∼25%, while good image display quality can be maintained.

This paper proposes a method to improve egomotion determination using latent semantic indexing for feature description and enhancing feature matching via an adaptive procedure. Binocular stereo images are processed using SURF to obtain feature descriptors, which are analyzed semantically using SVD to generate new feature descriptors. Matching criteria for image pairs are established based on epipolar geometry; the criteria are then adaptively adjusted between successive image pairs to achieve more reliable matches. Four types of experiments have been conducted to evaluate the proposed method. Experiment results demonstrated the effectiveness of the proposed approach in providing smaller numbers of initial match candidates with higher probability of correct matches, thus improving the efficiency and accuracy of ego-motion estimation.

Being expressed in immune cells, cytokine-inducible SH2 protein (CIS) and suppressors of cytokine signaling proteins, SOCS1, SOCS2 and SOCS3, can regulate cytokine signaling and immune responses. To evaluate the possible expressional dysregulation of CIS, SOCS1, SOCS2 and SOCS3 in systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) patients, the transcript levels of these genes in peripheral blood mononuclear cells (PBMCs) from SLE and RA patients were determined and statistically compared with those in PBMCs from normal individuals. It was found that SLE patients with the active disease activity significantly express higher CIS transcript levels than normal individuals and SLE patients with the inactive disease activity, whereas the difference in SOCS1, SOCS2 and SOCS3 transcript levels between normal individuals and SLE patients is not statistically significant. However, transcript levels of these CIS/SOCS genes in RA patients were not significantly different from those in normal individuals, except that treatment with a TNF-alpha-blocking agent in RA patients appears to enhance the CIS transcript expression, but down-regulates the SOCS2 transcript expression in PBMCs. These data suggest that CIS can serve as an SLE disease marker and may be involved in the pathogenesis of SLE, and that TNF-alpha may play an important role in the regulation of CIS and SOCS2 gene expression in PBMCs in vivo.

FOXO forkhead transcription factors play an important role in controlling lymphocyte activation and proliferation. To evaluate the possibility that FOXO transcriptional expression is dysregulated in systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) patients, we determined the quantities of FOXO1, FOXO3a, and FOXO4 transcripts in peripheral blood mononuclear cells (PBMCs) from normal controls as well as from SLE and RA patients. Results showed that FOXO1 and FOXO3a are dominant FOXO factors at the transcript level in PBMCs. Statistical analysis showed that the FOXO1 transcript levels in RA patients and in SLE patients with active disease activity were significantly lower than those in normal controls, and the FOXO1 transcript levels were inversely correlated with lupus disease activity. In contrast, the differences in FOXO3a and FOXO4 transcript levels between normal controls and patients were not significant. These data suggest that the transcriptional dysregulation in FOXO1 is possibly linked to the pathogenesis of SLE and RA.

This study proposes a 7T1R nonvolatile SRAM (nvSRAM) to 1) reduce store energy by using a single NVM device, 2) suppress DC-short current during restore operations through the use of a pulsed-overwrite (POW) scheme, and 3) achieves high restore yield by using a differentially supplied initialization (DSI) scheme. This initialization-and-overwrite (IOW) 7T1R nvSRAM improves breakeven-time (BET) by 6+x, compared to previous nvSRAMs. We fabricated a 16Kb IOW-7T1R nvSRAM using HfO x RRAM and a 90nm process. This represents the first ever silicon verified single-NVM nvSRAM macro. Measurements obtained in test-mode confirm that the proposed nvSRAM reduces store energy by 2x and restore energy by 94x, compared to 2R-based nvSRAMs.

To reach long operating time, Nonvolatile memory (NVM) has been applied to mobile systems in an attempt to reduce energy consumption through normally-off operation. The ReRAM device as a nonvolatile memory has shown promises in replacing current NVMs used in these applications, with orders of improvement in write speeds and read/write power than those offered by conventional Flash memories. Unfortunately, constraints related to endurance, read disturb, and resistance variations are still to be overcome for the wide usage of this device. This paper provides a review of the challenges and trends associated with the ReRAM circuits and applications used in mobile and cloud applications.

This study proposes an RC-filtered stress-decoupled (RCSD) 4T2R nonvolatile TCAM (nvTCAM) to 1) suppress match-line (ML) leakage current from match cells (I ML-M ), 2) reduce ML parasitic load (C ML ), 3) decouple NVM-stress from wordlength (WDL) and I ML-MIS . RCSD reduces NVM-stress by 7+x, and achieves a 4+x improvement in speed-WDL-capacity-product. A 128×32b RCSD nvTCAM macro was fabricated using HfO ReRAM and an 180nm CMOS. This paper presents the first ReRAM-based nvTCAM featuring the shortest (1.2ns) search delay (T SD ) among nvTCAMs with WDL≥32b.

Cytochrome P450 2C9 (CYP2C9) metabolizes dehydroepiandrosterone-sulfate (DHEA-S), but in elderly people the amount of DHEA-S remaining after CYP2C9 metabolization may be insufficient for optimal health. A prediction model, molecular docking, and molecular dynamics were used to screen the Traditional Chinese Medicine (TCM) database to determine molecular compounds that may inhibit CYP2C9. The candidate compounds apocynoside(I), 4-methoxymagndialdehyde, and prunasin have higher Dock Scores, and prediction bioactivity than warfarin (the control drug). The interaction between 4-methoxymagndialdehyde and CYP2C9 is more intense than with other TCM compounds, but the simulation is longer. In these compounds, apocynoside(I) and prunasin have a greater number of pathways for their flexible structure, but these structures create weak interactions. These candidate compounds, which are known to have antioxidation and hypolipidemic functions that have an indirect effect on the aging process, can be extracted from traditional Chinese medicines. Thus, these candidate compounds may become CYP2C9 inhibitors and play an important role in providing optimal health in the elderly.

This study demonstrates a new 7T2R nonvolatile SRAM (nvSRAM) with 3D ReRAM stacked structure for normally-off computing application. With this structure, the fully performance of SRAM can work well in active mode, and reduce the leakage current in power-off mode. High performance HfOx based ReRAM is used for high speed storage element and exhibits an instant-on characteristic. The present 7T2R nvSRAM cell includes a 1T2R RRAM (1 transistor/2 resistive memory) cell and a 6T SRAM circuit, which is low area penalty and achieve the nvSRAM function. The write margin is improved over 1.03x and 1.37x larger than that of 6T SRAM and 6T2R nvSRAM. The access time and read/write power consumption in 7T2R nvSRAM is better than that of 8T2R structure. Finally, a 16 Kb macro was fabricated with a 0.18 μm TSMC FEOL and ITRI BEOL. According to the measurement result, the VDDmin can be low down to 0.7 V and access time can be fast as 8.3 ns without pad delay. The data storage time is only 10 ns for SET and RESET in the ReRAM cell.

With its mixture of eastern and western cultures, Taiwan is full of amazing artistic performances, such as the Cloud Gate Dance Theatre [Yang 1998], which performs modern dance. Inspired by their fascinating performance, this study uses animation effects [White Rabbit Entertainment 2012] and parallel computing technology to bring a new vision to modern dance. All of this research was conducted at the render farm on Formosa series supercomputers [TOP500 Supercomputing Sites 2012] in Taiwan.

Several emerging nonvolatile memories (NVMs) including phase-change RAM (PCRAM) [1–3], MRAM [4–5], and resistive RAM (RRAM) [6–8] have achieved faster operating speeds than embedded Flash. Among those emerging NVMs, RRAM has advantages in faster write time, a larger resistance-ratio (R-ratio), and smaller write power consumption. However, RRAM cells have large cross-die and within-die resistance variations (R-variations) and require low read-mode bitline (BL) bias voltage (V BL-R ) to prevent read disturbance. This work proposes process/resistance variation-insensitive read schemes for embedded RRAM to achieve fast read speeds with high yields. An embedded mega-bit scale (4Mb), single-level-cell (SLC) RRAM macro with sub-8ns read-write random-access time is presented. Multi-level-cell (MLC) operation with 160ns write-ver-ify operation is demonstrated.

Scientific visualization is a powerful and popular technique and is very useful in many areas of science and engineering. It mainly uses computer graphics to view or explore three-dimensional (3D) objects, natural phenomenon, or numerical data. Virtual reality (VR) is an application of scientific visualization and is getting popular for applications associated with high-performance computing. To help Council of Agriculture, Taiwan (COA), promote the importance of farmland conservation, and to help elementary education authorities on basic agricultural, ecological education, in this study, a 3D virtual environment is developed to demonstrate the virtual farmland theme and associated temporal and spatial data and information. The system features (1) construction and visualization of the virtual farmland environment, (2) speedup and animation of the embedded objects, (3) multimedia effects, (4) display of the numerical analysis results of crops growing affected by nitrogen in land simulated by 2DSOIL model, sand (5) display of the numerical analysis results of groundwater flow and nitrogen transport in the farmland simulated by GMS-FEMWATER. Several computer graphics accelerating schemes, such as space partition, view frustum culling (VFC), billboard, and collision detection, were used to speedup the rendering. Users can immerse, navigate and interact with objects in the 3D virtual farmland ecology environment.

We provide a Web-based interface for numerical data representation based on WAN (wide area networking). Furthermore, we provide communication facilities for each module and make the package development and maintenance easy. The system, Visual WIS (Water Information System), has been developed by JDK1.2. By incorporating a Java-based tool into GMS (Groundwater Modeling System), many groundwater problems can be solved and demonstrated on the World Wide Web. A procedure to adjust the control functions, proposed by Yen and Kuo (1998), is adopted for generating curvilinear grid nodes to the specified locations of the study cases.