应用程序属性感知的Yarn资源调度模型研究

Hadoop应用程序存在计算密集属性和调度时间属性,但是Hadoop大数据平台集成的第二代资源管理器Yarn内置的三种资源调度器无法将相同属性的应用程序均衡分配到计算节点上,导致部分节点负载过高,出现严重的计算任务长尾效应。文中提出了一种应用程序属性感知的Yarn负载均衡调度模型——APB Scheduler。APB Scheduler自动感知应用程序属性,将相同属性应用程序的Container按照动态资源计划均衡分配到集群计算节点上,并使用NSGA-Ⅲ算法完成最优分配方案计算。实验结果表明,APB Scheduler解决了相同属性应用程序的Container分配倾斜问题,大幅提升了集群的性能和稳定性。

Dual-Ion Co-Regulation System Enabling High-Performance Electrochemical Artificial Yarn Muscles with Energy-Free Catch States

Artificial yarn muscles show great potential in applications requiring low-energy consumption while maintaining high performance. However, conventional designs have been limited by weak ion-yarn muscle interactions and inefficient “rocking-chair” ion migration. To address these limitations, we present an electrochemical artificial yarn muscle design driven by a dual-ion co-regulation system. By utilizing two reaction channels, this system shortens ion migration pathways, leading to faster and more efficient actuation. During the charging/discharging process, PF_6~- ions react with carbon nanotube yarn, while Li~+ ions react with an Al foil. The intercalation reaction between PF_6~- and collapsed carbon nanotubes allows the yarn muscle to achieve an energy-free high-tension catch state. The dual-ion coordinated yarn muscles exhibit superior contractile stroke, maximum contractile rate, and maximum power densities, exceeding those of “rocking-chair” type ion migration yarn muscles. The dual-ion co-regulation system enhances the ion migration rate during actuation, resulting in improved performance. Moreover, the yarn muscles can withstand high levels of isometric stress, displaying a stress of 61 times that of skeletal muscles and 8 times that of “rocking-chair” type yarn muscles at higher frequencies. This technology holds significant potential for various applications, including prosthetics and robotics.

Yarn Quality Prediction for Small Samples Based on AdaBoost Algorithm

In order to solve the problems of weak prediction stability and generalization ability of a neural network algorithm model in the yarn quality prediction research for small samples,a prediction model based on an AdaBoost algorithm(AdaBoost model) was established.A prediction model based on a linear regression algorithm(LR model) and a prediction model based on a multi-layer perceptron neural network algorithm(MLP model) were established for comparison.The prediction experiments of the yarn evenness and the yarn strength were implemented.Determination coefficients and prediction errors were used to evaluate the prediction accuracy of these models,and the K-fold cross validation was used to evaluate the generalization ability of these models.In the prediction experiments,the determination coefficient of the yarn evenness prediction result of the AdaBoost model is 76% and 87% higher than that of the LR model and the MLP model,respectively.The determination coefficient of the yarn strength prediction result of the AdaBoost model is slightly higher than that of the other two models.Considering that the yarn evenness dataset has a weaker linear relationship with the cotton dataset than that of the yarn strength dataset in this paper,the AdaBoost model has the best adaptability for the nonlinear dataset among the three models.In addition,the AdaBoost model shows generally better results in the cross-validation experiments and the series of prediction experiments at eight different training set sample sizes.It is proved that the AdaBoost model not only has good prediction accuracy but also has good prediction stability and generalization ability for small samples.

The Influence of Elimination of Roving Frame on Ring Spun Cotton Yarn Characteristics

Generally, ring spun yarns are manufactured from roving which is produced by roving frame. In this paper, an experiment has been done producing ring spun cotton yarn directly from finisher drawn sliver eliminating the roving frame. Total 3 types of yarn with the various linear density of 8 Ne, 10 Ne & 12 Ne were produced using a roving frame and without using a roving frame. In the next step, physical and mechanical properties of those yarns including unevenness, imperfections, hairiness & tenacity were investigated. The result showed that ring spun cotton yarns produced from sliver exhibited inferior physical and mechanical properties compared with samples from the conventional ring spinning system.

Scalable Core–Sheath Yarn for Boosting Solar Interfacial Desalination Through Engineering Controllable Water Supply

Tailoring water supply to achieve confined heating has proven to be an effective strategy for boosting solar interfacial evaporation rates.However,because of salt clogging during desalination,a critical point of constriction occurs when controlling the water rate for confined heating.In this study,we demonstrate a facile and scalable weaving technique for fabricating core-sheath photothermal yarns that facilitate controlled water supply for stable and efficient interracial solar desalination.The core-sheath yarn comprises modal fibers as the core and carbon fibers as the sheaths.Because of the core-sheath design,remarkable liquid pumping can be enabled in the carbon fiber bundle of the dispersed superhydrophilic modal fibers.Our woven fabrics absorb a high proportion(92%)of the electromagnetic radiation in the solar spectrum because of the weaving structure and the carbon fiber sheath.Under one-sun(1 kW·m^(-2))illumination,our woven fabric device can achieve the highest evaporation rate(of 2.12kg·m^(-2)·h^(-1) with energy conversion efficiency:93.7%)by regulating the number of core-sheath yarns.Practical application tests demonstrate that our device can maintain high and stable desalination performance in a 5 wt%NaCl solution.

Study on the Presence of Filament Yarn in Jamdani Saree in Bangladesh

Jamdani weaving is one of the oldest heredities of Bangladesh. From the beginning 100% cotton yarn was used to produce high quality jamdani saree. The weavers were the finest with weaving skills. Higher yarn count yarns were used to weave the jamdani saree. In course of time at present manmade fibres are also used to produce jamdani saree. The use of filament yarn may have eased the manufacturing difficulties, but the jamdani saree is missing its originality without 100% cotton. In this project, random jamdani saree sample was collected to identify the fibre composition. Samples of filament were also collected from the manufacturer and tested. It was evident that instead of cotton yarn in warp and weft silk and polyester filament yarn were used.

Are yarn quality prediction tools useful in the breeding of high yielding and better fibre quality cotton(Gossypium hirsutum L.)?

Results The population had large variations for lint yield,fibre properties,predicted yarn properties,and composite fibre quality values.Lint yield with all fibre quality traits was not correlated.When the selection was conducted first to keep those with improved fibre quality,and followed for high yields,a large proportion in the resultant populations was the same between selections based on Cottonspec predicted yarn quality and HVI-measured fibre properties.They both exceeded the selection based on FQI and Background The approach of directly testing yarn quality to define fibre quality breeding objectives and progress the selection is attractive but difficult when considering the need for time and labour.The question remains whether yarn prediction tools from textile research can serve as an alternative.In this study,using a dataset from three seasons of field testing recombinant inbred line population,Cottonspec,a software developed by the Commonwealth Scientific and Industrial Research Organisation(CSIRO)for predicting ring spun yarn quality from fibre properties measured by High Volume Instrument(HVI),was used to select improved fibre quality and lint yield in the population.The population was derived from an advanced generation inter-crossing of four CSIRO conventional commercial varieties.The Cottonspec program was able to provide an integrated index of the fibre qualities affecting yarn properties.That was compared with selection based on HVI-measured fibre properties,and two composite fibre quality variables,namely,fibre quality index(FQI),and premium and discount(PD)points.The latter represents the net points of fibre length,strength,and micronaire based on the Premiums and Discounts Schedule used in the market while modified by the inclusion of elongation.PD points.Conclusions The population contained elite segregants with improved yield and fibre properties,and Cottonspec predicted yarn quality is useful to effectively capture these elites.There is a need to further develop yarn quality prediction tools through collaborative efforts with textile mills,to draw better connectedness between fibre and yarn quality.This connection will support the entire cotton value chain research and evolution.

Detection of Residual Yarn in Bobbin Based on Odd Partial Gabor Filter and Multi-Color Space Hierarchical Clustering

In an automatic bobbin management system that simultaneously detects bobbin color and residual yarn,a composite texture segmentation and recognition operation based on an odd partial Gabor filter and multi-color space hierarchical clustering are proposed.Firstly,the parameter-optimized odd partial Gabor filter is used to distinguish bobbin and yarn texture,to explore Garbor parameters for yarn bobbins,and to accurately discriminate frequency characteristics of yarns and texture.Secondly,multi-color clustering segmentation using color spaces such as red,green,blue(RGB)and CIELUV(LUV)solves the problems of over-segmentation and segmentation errors,which are caused by the difficulty of accurately representing the complex and variable color information of yarns in a single-color space and the low contrast between the target and background.Finally,the segmented bobbin is combined with the odd partial Gabor’s edge recognition operator to further distinguish bobbin texture from yarn texture and locate the position and size of the residual yarn.Experimental results show that the method is robust in identifying complex texture,damaged and dyed bobbins,and multi-color yarns.Residual yarn identification can distinguish texture features and residual yarns well and it can be transferred to the detection and differentiation of complex texture,which is significantly better than traditional methods.

Performance Improvement through Novel Adaptive Node and Container Aware Scheduler with Resource Availability Control in Hadoop YARN

The default scheduler of Apache Hadoop demonstrates operational inefficiencies when connecting external sources and processing transformation jobs.This paper has proposed a novel scheduler for enhancement of the performance of the Hadoop Yet Another Resource Negotiator(YARN)scheduler,called the Adaptive Node and Container Aware Scheduler(ANACRAC),that aligns cluster resources to the demands of the applications in the real world.The approach performs to leverage the user-provided configurations as a unique design to apportion nodes,or containers within the nodes,to application thresholds.Additionally,it provides the flexibility to the applications for selecting and choosing which node’s resources they want to manage and adds limits to prevent threshold breaches by adding additional jobs as needed.Node or container awareness can be utilized individually or in combination to increase efficiency.On top of this,the resource availability within the node and containers can also be investigated.This paper also focuses on the elasticity of the containers and self-adaptiveness depending on the job type.The results proved that 15%–20%performance improvement was achieved compared with the node and container awareness feature of the ANACRAC.It has been validated that this ANACRAC scheduler demonstrates a 70%–90%performance improvement compared with the default Fair scheduler.Experimental results also demonstrated the success of the enhancement and a performance improvement in the range of 60%to 200%when applications were connected with external interfaces and high workloads.

Spark on Yarn模式的电信大数据处理平台

针对电信大数据处理系统中存在的问题,文章提出一种基于Spark on Yarn模型的SY-TPP。在SY-TPP平台上,应用Hadoop2.0 Yarn标准,并利用Spark分布式存储技术,将SY-TPP系统的数据在内存中进行集中处理。以分级聚类算法为案例,对SY-TPP平台的开发过程进行了详细的分析。实验结果表明,TPP平台上的GB级用户可以在半个工作日内完成数据处理,而32个实体节点的SYTPP系统的速度比相同配置下的Map Reduce平台提高了10.25倍。

纯亚麻短麻色纺纱的生产及成纱性能

针对当前亚麻短麻色纺纱少且短麻利用率不高等问题,采用梳棉试样机对化学脱胶后的亚麻短麻纤维进行2次梳理开松,再经染色处理、给油加湿焖包处理,最后在棉纺设备上进行试纺。通过优化纺纱各工序的工艺参数配置,成功纺制出41.7 tex纯亚麻短麻色纺纱,测试了其成纱质量指标,并与41.7 tex亚麻湿纺染色纱、41.7 tex纯亚麻纱线染色纱进行性能对比。结果表明:在棉纺设备上所纺制的41.7 tex纯亚麻短麻色纺纱的主要成纱指标达到了T/CNTAC60—2020《干纺环锭纺纯亚麻本色纱》优等纱质量标准;其断裂强度、强力CV值、断裂伸长率等均介于同线密度的纯亚麻湿纺染色纱和纯亚麻纱线染色纱之间,可部分替代纯亚麻湿纺染色纱用于后续织造工艺。

基于改进遗传算法的细纱接头路径指引方法

为了提高细纱车间整体断头的接头效率,在细纱单锭监测系统响应到细纱断头信息的条件下,建立以整体车间断头接头路径最优化为目标函数的细纱接头路径指引模型,并运用Python编程语言分别实现对贪心算法、遗传算法和改进遗传算法接头路径模型的求解,结合传统巡回式接头路径与3种算法在效率、运行速度和解的质量上的仿真测试,对比验证改进遗传算法用于细纱接头路径指引方法的可行性。试验结果表明:相较于传统巡回式路径,改进遗传算法平均效率提升了11.8%,且解均优于其他两种算法;相较于遗传算法,改进遗传算法的响应时间缩短较多,平均为1.25 s,且波动较小,时间效率提升了80.3%。认为:改进遗传算法在效率、运行速度和解的质量上都优于其他算法,在细纱接头领域具有较高的应用优势,能够显著提高接头效率和降低成本。

纬编电热针织物设计和性能评价

为了探究金属长丝作为电加热元件在纬编电热针织物中的应用,通过赛络菲尔纺工艺开发了棉/铜镁合金和棉/不锈钢复合导电纱,并与碳纤维进行比较,采用罗纹空气层加衬纬组织制作了电加热元件并联的3种电热织物,并对导电纱和纬编织物电热性能进行实测及数据拟合。结果显示:复合导电纱达到导电纱A类产品要求;在12 V电压时,棉/铜镁合金导电纱表面温度最高可达40.9℃;金属复合导电纱电热织物升温和时间、电压有关,电热织物的表面平衡温度与输入功率呈线性关系,温度达45℃以上时,电热织物电阻变化率低于2%,热稳定性好;相同条件下,棉/铜镁合金织物升温速度最快,输入功率最大,在4 V电压下最高温度达49.8℃;采用赛络菲尔复合导电纱开发罗纹空气层加衬纬的纬编电加热织物具有一定的可行性。

基于自适应Loess的纱线张力信号处理方法

为保持针织系统中纱线张力的稳定并解决纱线张力检测过程中的噪声干扰问题,提出了一种基于自适应Loess的纱线张力信号优化方法。通过信号特征分析,将纱线张力结果划分为周期性信号、奇异点信号、高频干扰信号和低频耦合干扰信号。为消除这些干扰信号,提出了相应的数据处理方法:限幅滤波去除奇异点信号、低通滤波消除高频干扰信号以及改进的Loess方法去除低频耦合干扰信号。在3种不同情况的张力信号上验证了该方法的可靠性。实验结果显示,自适应Loess算法能有效平滑张力波动,提升信噪比分别达25.014%、27.661%、25.276%,证明了该方法在不同情况下的有效性和稳定性,为针织系统中纱线张力的稳定性和生产效率的提升提供了可行的解决方案。

基于逆投影变换的纱筒纱线余量检测算法

纱筒纱线余量检测是纺织行业自动化生产过程中的重要一环,针对目前纱线余量检测算法检测精度低的问题,提出一种基于三点逆投影变换模型的纱筒纱线余量检测技术。根据实际生产过程中相机与纱筒的位置关系,建立三点逆投影变换模型进行初步矫正。为得到更好的图像矫正结果,设置补偿矩阵并根据内外筒位置信息优化变换矩阵,最后将逆投影变换后的图像按照改进的椭圆坐标变换模型展开为矩形,根据余量计算准则计算得到纱筒纱线余量。在自主搭建的检测平台上进行实验,结果表明,本文算法检测精度在5 mm以内,基本满足实际生产要求,可为纺织产业自动化生产提供依据。

基于自抗扰控制的圆纬机恒张力输纱控制系统设计

针对当前圆纬机积极式送纱方式中存在的纱线张力及输送速度不可任意调节、纱线张力波动大等问题,提出了一种可为圆纬机输送恒定张力纱线的输纱控制系统。构建了以无刷直流电动机速度环为内环、以纱线张力为外环的纱线恒张力双闭环反馈控制系统,通过自抗扰控制技术获取纱线实时动态指标,并对无刷直流电动机目标转速进行调整,最后通过磁场定向控制技术驱动无刷直流电动机。详细阐述了纱线张力的形成机制、纱线恒张力输送控制方案、系统硬件设计及控制算法。输纱测试表明,该控制系统可实现多种类型纱线的恒定张力输送。其中,棉纱张力波动标准差为1.5 cN(300 m/min测试条件),氨纶包芯纱的张力波动标准差为0.33 cN(240 m/min测试条件)。输纱速度的调节范围为0~600 m/min,纱线张力的控制范围为2.0~50.0 cN。

基于印花效果的四色经提花面料产品研发

为研发具有印花效果的四色经提花面料新产品,对产品的花稿设计、纱线配置和编织工艺3方面内容进行了研究与实践。通过设计合适的花稿图案,选用5.56tex/36f的加捻涤纶低弹色纱,色彩分别为红、蓝、黄、黑4种颜色的纱线作为经纱,8.33tex/72f的白色或黑色涤纶色纱作为纬纱,运用重经组织结构使这4种不同颜色经纱形成不同的浮长,利用经纱混色配置的方式实现花稿图案的丰富色彩,研发出色彩丰富且立体感强、面密度轻、环保性能好、具有印花效果的轻薄型提花面料。

雅赛尔11.8 tex弹力包芯纱生产实践

为了提高雅赛尔纤维弹力纱纺纱质量,以纺雅赛尔11.8 tex弹力包芯纱为例,介绍雅赛尔、氨纶弹力丝的性能指标及工艺流程;详细分析纺纱各生产工序采取的技术措施、工艺配置及注意事项;通过对细纱、筒纱质量指标的测试,说明雅赛尔11.8 tex弹力包芯纱成纱指标良好、质量稳定。指出:提高雅赛尔纤维品种强力、减少其成纱毛羽、控制氨纶丝包覆质量是保证成纱质量的关键;生产过程中,要优选各工序工艺,明确控制要点及参数,合理配置相关器材和车间温湿度,加强各工序过程控制和设备维护保养。

基于YARN集群的计算加速部件扩展支持

以GPU和Intel MIC为代表的计算加速部件已在科学计算、图形图像处理等领域得到了广泛的应用,其在基于云平台的高性能计算及大数据处理等方向也具有广泛的应用前景.YARN是新一代Hadoop分布式计算框架,其对计算资源的分配调度主要针对CPU,缺少对计算加速部件的支持.在YARN中添加计算加速部件需要解决多个难点,分别是计算加速部件资源如何调度以及异构节点间如何共享问题、多个任务同时调用计算加速部件而引起的资源争用问题和集群中对计算加速部件的状态监控与管理问题.为了解决这些问题,提出了动态节点捆绑策略、流水线式的计算加速部件任务调度等,实现了YARN对计算加速部件的支持,并通过实验验证了其有效性.

Yarn架构下基于GA的Web日志挖掘技术

提出了一种面向TB级别日志文件挖掘需求的日志挖掘技术。采用MapReduce编程模型设计并实现了一种基于子种群联姻策略的GA,并部署到Yarn架构上,使Yarn架构与GA有效结合。在算法Partition阶段,采用Round-Robin策略代替原有的哈希方法,使各个子种群基因更加趋向均匀分布,增加了子种群收敛效率和结果准确性,同时均衡各个节点运行负载,提高了系统性能。经测试,应用该技术的挖掘结果平均准确度达到93%以上,效率提升接近33%。