JAVA垃圾收集器算法分析及垃圾收集器的运行透视

本文分析了JVM垃圾收集器所使用的多种收集算法,讨论如何通过命令行参数和终止化方法,透视JAVA自动垃圾收集器的运行,以理解JVM垃圾自动回收机制.程序开发人员应当使用finalize方法明确释放对象资源,并通过GC请求JVM执行垃圾收集,使得JVM尽可能多地回收内存供程序使用.     

一种改进的增量式JVM垃圾收集算法

非增量式Java虚拟机(JVM)垃圾回收算法的内存开销较大。为此,提出一种基于栈式分配策略的JVM增量式垃圾收集算法。对Java栈帧进行改造使其支持存储对象,改进增量式收集器中堆空间的划分、引用跟踪方式,以减少垃圾收集带来的不确定性暂停。实验结果表明,该算法能有效减少暂停的频率和时长,提高运行速度。     

JVM垃圾收集算法分析及选择策略

JVM 的垃圾收集算法对应用程序的性能有着非常重要的影响．有些时候甚至成为应用程序的性能瓶颈,因此Java学习者有必要了解这一内部机制。阐述JVM垃圾收集的机制、原理。着重分析几种常用的垃圾收集算法,给出垃圾收集算法的选择策略。     

Java堆的管理——垃圾回收机制的算法分析与研究

分析了Java垃圾回收机制的特点,分析和研究了JVM中典型的垃圾收集算法,并阐述了命令行参数和Finalize方法时Java垃圾回收的具体运行。     

基于垃圾收集的Java程序性能改善方法*

JVM中堆的默认配置有时并不适用于大型程序,这会使系统在垃圾收集上耗费过多的时间,进而导致程序性能下降。简要分析了HotSpot JVM垃圾收集机制,然后结合已有研究,以一个网管程序为例,提出了通过调整垃圾收集提高程序运行效能的方法。     

JVM垃圾收集器的性能分析及编码建议

阐述了JVM垃圾收集的机制、原理,着重分析了JVM垃圾收集机制可能对系统性能带来的负面影响,最后给出了相应的Java编码建议。     

Java虚拟机垃圾收集器的性能分析与调节

垃圾收集器是Jaya虚拟机(JVW)的核心组成部分之一，对Java虚拟机的性能有非常重要的影响。本文将介绍GC的工作原理以及对象回收算法，重点介绍JVM的分段回收技术；剖析JVM自带的GC性能分析工具；阐述如何通过命令行参数调节GC的运行，提高GC的效率。     

面向大数据处理框架的JVM优化技术综述

当前,以Hadoop、Spark为代表的大数据处理框架,已经在学术界和工业界被广泛应用于大规模数据的处理和分析.这些大数据处理框架采用分布式架构,使用Java、Scala等面向对象语言编写,在集群节点上以Java虚拟机(JVM)为运行时环境执行计算任务,因此依赖JVM的自动内存管理机制来分配和回收数据对象.然而,当前的JVM并不是针对大数据处理框架的计算特征设计的,在实际运行大数据应用时经常出现垃圾回收(GC)时间长、数据对象序列化和反序列化开销大等问题.在一些大数据场景下, JVM的垃圾回收耗时甚至超过应用整体运行时间的50%,已经成为大数据处理框架的性能瓶颈和优化热点.对近年来相关领域的研究成果进行了系统性综述:(1)总结了大数据应用在JVM中运行时性能下降的原因;(2)总结了现有面向大数据处理框架的JVM优化技术,对相关优化技术进行了层次划分,并分析比较了各种方法的优化效果、适用范围、使用负担等优缺点;(3)探讨了JVM未来的优化方向,有助于进一步提升大数据处理框架的性能.     

浅析JVM性能规范

Java语言有着独特的工作原理,这是由于Java引入的虚拟机(JVM)以及垃圾回收机制等功能所带来的.一切JavaEE应用都是基于JVM的,这些工作原理具体是如何工作,带来了怎样的便利,本文做出了比较详细的解答.     

适合于分布并行计算的JVM—Lkaffe的设计和实现

介绍了Alpha工作站Liunx操作系统上一种能实现真正并发的JVM－Lkaffe的设计和实现,并着重介绍了它的线程实现机制,垃圾回收实现机制和异常处理实现机制。     

Gambling content in Facebook games: A common phenomenon?

Some Facebook games are offered by developers who also offer gambling games, possibly indicating that gambling content (GC) could be found in their Facebook games. This study measures the presence of GC in Facebook games and documents their presentation. It verifies whether GC is more present in games offered by developers offering gambling games as well. The 100 most popular Facebook games were played for 10 min and recorded for content analysis purposes. GC was detected and classified into standard gambling simulation, non-standard gambling simulation, and gambling references. The results indicate that 54% of Facebook games present GC. The GC most often used alluded to slot machines (22%). Facebook games which offer GC are not associated with gambling game developers. However, when gambling references are found, they were most often in games offered by gambling developers. GC as a medium for familiarization with gambling is discussed.     

Polyphenol-modified glassy carbon electrodes for copper detection

The present work reports the preparation of electrochemically polymerized flavone - luteolin and flavonol - kaempferol modified glassy carbon (GC) electrodes (PolyLut/GC and PolyKae/GC, respectively). Electrochemical polymerization was performed by electrochemical oxidation of luteolin and kaempferol by potential cycling in aqueous media. Cyclic voltammograms of luteolin on the GC electrode indicated one clear oxidation peak at +475 mV, which can be assigned to the oxidation of 3′-hydroxyl and 4′-hydroxyl groups in the B-ring of the luteolin molecule. The cyclic voltammograms of kaempferol on the GC electrode contained two oxidation peaks, one at about +390 mV, which is assigned to the oxidation of 4′-hydroxyl and 3-hydroxyl groups of the B-ring and C-ring of the kaempferol molecule, and second oxidation peak at about +710 mV, which is assigned to the 7-hydroxyl group of the A-ring. The interaction of PolyLut/GC and PolyKae/GC electrodes with copper(II) (Cu(II)) ions were investigated by differential pulse voltammetry (DPV). It was determined that PolyLut/GC and PolyKae/GC electrodes showed sensitivity towards Cu(II) with good reproducibility and stability of analytical signal. The effect of the interfering ions on the voltammetric measurements of Cu(II) was examined.     

1,10-Phenanthroline modified glassy carbon electrode for voltammetric determination of cadmium(II) ions

This study reports the electrochemical modification of glassy carbon (GC) electrode surface with the electro-polymerized form of 1,10-phenanthroline monohydrate (PMH), the characterization of this polyphenanthroline modified electrode (PPMH/GC) and the electroanalytical application suitable for the determination of Cd(II) ions. The PPMH/GC electrode was characterized by cyclic voltammetry, chronoamperometry and atomic force microscopy and formation of polyphenanthroline layer grafted to surface of GC electrode was evidenced. Selectivity of PPMH/GC electrode towards heavy metal ions was investigated by square wave voltammetry. The PPMH/GC electrode was found to be suitable for selective determination of Cd(II) in the solutions containing the mixture of heavy metal ions and showed high stability and reproducibility. The analytical methodology was successfully applied for monitoring the toxic metal ions in real samples.     

Achieving middleware execution efficiency: hardware-assisted garbage collection operations

Although virtualization technologies bring many benefits to cloud computing environments, as the virtual machines provide more features, the middleware layer has become bloated, introducing a high overhead. Our ultimate goal is to provide hardware-assisted solutions to improve the middleware performance in cloud computing environments. As a starting point, in this paper, we design, implement, and evaluate specialized hardware instructions to accelerate GC operations. We select GC because it is a common component in virtual machine designs and it incurs high performance and energy consumption overheads. We performed a profiling study on various GC algorithms to identify the GC performance hotspots, which contribute to more than 50% of the total GC execution time. By moving these hotspot functions into hardware, we achieved an order of magnitude speedup and significant improvement on energy efficiency. In addition, the results of our performance estimation study indicate that the hardware-assisted GC instructions can reduce the GC execution time by half and lead to a 7% improvement on the overall execution time.     

全局最优的邻域估计

为了实现全局最优的快速搜索,对全局最优所处的范围进行预估非常关键.受沙盘重心的力平衡关系启发,提出基于重心定位的全局最优邻域估计方法.利用空间变换技术,将寻优空间的重心＂移＂至全局最优附近;在重心逐次＂移＂动过程中,以重心为几何中心,构造一个范围不断缩小的全局最优邻域.函数测试和工程应用表明,所提出方法能以较小的种群和迭代次数确定全局最优的有效邻域范围.     

Fabrication of glass-like carbon molds to imprint on glass materials by MEMS processing technologies

We processed a precise relief structure on the surface of a glass-like carbon (GC) substrate by applying micro-electro-mechanical-systems (MEMS) technologies, and made a high temperature resistant mold for thermal imprinting on glass materials. An attractive feature of GC is its chemical stability at high temperatures (above 1,000 °C). The down side is its brittleness that makes microfabrication with GC a difficult task. We investigated to find if photolithography combined with reactive-ion-etching (RIE), which are generally used in MEMS fabrication, could be applied for the fabrication of GC molds. In our work with the RIE process, we made masking layers using Au and a positive-tone photoresist. By taking advantage of the difference between the etching rates of the masking materials and GC, we fabricated convex mold patterns with vertical and curved sidewalls. From the experimental results imprinted on Pyrex glass and on quartz, the practicability of using both kinds of GC molds appeared to be quite promising. We believe that in the near future these techniques will be successfully applied in the fabrication of large-size GC molds.     

Java垃圾收集的机制及调优

阐述Java垃圾收集的作用，归纳垃圾收集对Java应用性能的主要负面影响，分析垃圾收集的机制与算法，对垃圾收集提出调优办法。     

Using the red-near infrared spectral to estimate ground cover based on vegetative indices

Rapid and accurate estimation of Ground Cover (GC) at regional and global scales for agricultural management application is only possible by using remote sensing (RS). In this study, two Vegetation Indices (VIs) including the Perpendicular Vegetation Index (PVI) and Normalized Difference Vegetation Index (NDVI) were used for estimating GC. Since the parameters of the bare soil line have an important role in calculating GC based on PVI, this line was extracted based on the red-NIRmin (minimum near infrared) method with different intervals (0.0001, 0.0005, and 0.0010). In addition to traditional statistics such as Root Mean Square Error (RMSE), the sensitivity analysis (S) was also used to sharpen the accuracy of the models' estimations. The results indicated that the PVI-based method, in contrast to the NDVI-based approach, had a better performance in estimating GC of wheat. The highest correlation between the observed GC and the estimated GC based on PVI method was achieved in interval length of 0.0005 (R2 = 0.91) with RMSE equal to 8.82. This regression line (GCEST = -3.47 + 0.96 GCOBS) was not significantly different from the 1:1 line. As expected, the best estimation was achieved when the sensitivity of estimated GC based on PVI (length of the interval: 0.0005) was almost constant and low compared to the other models.     

全程优化的固态硬盘垃圾回收方法

由于NAND闪存的固有限制,写前擦除和擦除粒度较大,基于NAND Flash的固态硬盘（SSD）需要执行垃圾回收以重用失效页。然而垃圾回收带来的高开销会显著降低SSD的性能,也会直接影响SSD的寿命。特别是对于频繁使用的有数据碎片的SSD,垃圾回收带来的性能下降问题将更为严重,现有的垃圾回收（GC）算法各自侧重垃圾回收操作的某个步骤,并没有给出全面考虑各步骤对整体影响的综合方案。针对该问题,在详细剖析垃圾回收过程的基础上,提出了一种全程优化的垃圾回收方法WPO-GC,在数据初始放置、垃圾回收目标块的选择、有效数据的迁移、触发回收的时间点以及中断处理方式上,尽可能全面地考虑各步骤对SSD正常读写请求和寿命的影响。通过开源模拟器SSDsim上的WPO-GC的有效性验证表明,同典型GC算法相比,WPO-GC可以减少SSD读请求延迟20%~40%和写请求延迟17%~40%,均衡磨损近30%。     

Memory power optimization of Java-based embedded systems exploiting garbage collection information

Nowadays, Java is used in all types of embedded devices. For these memory-constrained systems, the automatic dynamic memory manager (Garbage Collector or GC) has been always a key factor in terms of the Java Virtual Machine (JVM) performance. Moreover, in current embedded platforms, power consumption is becoming as important as performance. Thus, in this paper we present an exploration, from an energy viewpoint, of the different possibilities of memory hierarchies for high-performance embedded systems when used by state-of-the-art GCs. This is a starting point for a better understanding of the interactions between the Java applications, the memory hierarchy and the GC.Hence, we subsequently present two techniques to reduce energy consumption on Java-based embedded systems, based on exploiting GC information. The first technique uses GC execution behavior to reduce leakage energy consumption taking advantage of the low-power mode of actual multi-banked SDRAM memories and it is intended for generational collectors. This technique can achieve a reduction up to 50% of SDRAM memory leakage.The second technique involves the inclusion of a software-controlled (scratch-pad) memory that stores GC instructions under the JVM control to reduce the active energy consumption and also improve the performance of the target embedded system and it is aimed at all kind of garbage collectors. For this last technique we have experimented with two different approaches for selecting the GC code to be stored in the scratchpad memory: one static and one dynamic. Our experimental results show that the proposed dynamic scratchpad management approach for GCs enables up to 63% energy consumption reduction and 25% performance improvement during the collector phase, which means, in terms of JVM execution, a global reduction of 29% and 17% for energy and cycles, respectively.Overall, this work outlines that the key for an efficient low-power implementation of Java Virtual Machines for high-performance embedded systems is the synergy between the GC choice, the memory architecture tuning, and the inclusion of power management schemes controlled by the JVM, exploiting knowledge of the GC behavior.