Optimizing I/O server placement for parallel I/O on switch-based irregular networks

In this paper, we study I/O server placement for optimizing parallel I/O performance on switch-based clusters, which typically adopt irregular network topologies to allow construction of scalable systems with incremental expansion capability. Finding optimal solution to this problem is computationally intractable. We quantified the number of messages travelling through each network link by a workload function, and developed three heuristic algorithms to find good solutions based on the values of the workload function. The maximum-workload-based heuristic chooses the locations for I/O nodes in order to minimize the maximum value of the workload function. The distance-based heuristic aims to minimize the average distance between the compute nodes and I/O nodes, which is equivalent to minimizing average workload on the network links. The load-balance-based heuristic balances the workload on the links based on a recursive traversal of the routing tree for the network. Our simulation results demonstrate performance advantage of our algorithms over a number of algorithms commonly used in existing parallel systems. In particular, the load-balance-based algorithm is superior to the other algorithms in most cases, with improvement ratio of 10 to 95% in terms of parallel I/O throughput.     

A disk I/O optimized system for concurrent graph processing jobs

In order to analyze and process the large graphs with high cost efficiency, researchers have developed a number of out-of-core graph processing systems in recent years based on just one commodity computer. On the other hand, with the rapidly growing need of analyzing graphs in the real-world, graph processing systems have to efficiently handle massive concurrent graph processing (CGP) jobs. Unfortunately, due to the inherent design for single graph processing job, existing out-of-core graph processing systems usually incur unnecessary data accesses and severe competition of I/O bandwidth when handling the CGP jobs. In this paper, we propose GraphCP, a disk I/O optimized out-of-core graph processing system that efficiently supports the processing of CGP jobs. GraphCP proposes a benefit-aware sharing execution model to share the I/O access and processing of graph data among the CGP jobs and adaptively schedule the graph data loading based on the states of vertices, which efficiently overcomes above challenges faced by existing out-of-core graph processing systems. Moreover, GraphCP adopts a dependency-based future-vertex updating model so as to reduce disk I/Os in the future iterations. In addition, GraphCP organizes the graph data with a Source-Sorted Sub-Block graph representation for better processing capacity and I/O access locality. Extensive evaluation results show that GraphCP is 20.5× and 8.9× faster than two out-of-core graph processing systems GridGraph and GraphZ, and 3.5× and 1.7× faster than two state-of-art concurrent graph processing systems Seraph and GraphSO.     

Mach的I／O系统

ＭａｃｈＩ／Ｏ系统采用了和ＵＮＩＸ完全不同的概念和结构。Ｍａｃｈ设备管理围绕端口和存储对象这两个Ｍａｃｈ基本概念进行,提供了方便的ＰＲＣ用户界面。Ｍａｃｈ３．０的Ｉ／Ｏ系统新进展将Ｉ／Ｏ管理作为用户Ｓｅｒｖｅｒ对待。本文介绍了Ｍａｃｈ设备管理机制、“设备独立”的驱动程序的构造原理及其例子、在用户空间对设备的直接控制方法和引进新概念后设备管理的性能情况。     

A schedule of join operations to reduce I/O cost in spatial database systems

In this paper, we propose a graph-based cluster-sequencing method to minimize the I/O cost in spatial join processing. We first define the maximum overlapping (MO) order in a graph, proving that the problem of finding an MO order in a graph is NP-complete. Then, we propose an algorithm to find an approximation to MO order in a graph. We also prove that the approximation to MO order obtained from our method is close to the optimal result. Simulations have been conducted to demonstrate the saving of I/O cost in spatial join by using our method.     

Assessing determinants of desirable ES impact on end-user jobs

To increase the likelihood of experts systems' success it is useful to identify the more important determinants to focus managerial attention and resources. Seven major determinants of expert system success, here defined as desirable impact of the system on users' jobs, have been identified from the literature. DuPont's collection of expert systems (ES) provided the sample of 114 applications used in this study. The results corroborate much of what has been proposed in the literature, except for user characteristics which have been found to have no significant relationship to the measure of success. Based on the results, some recommendations are made to expert system development managers, including: (1) besides the recommendations already proposed in the literature, managers should deliberately seek to deal with important, large, labour intensive and complex problems which are important to the organization; (2) the selection of a development shell with the characteristics described in this study, and which matches the problem at hand is an important determinant of success; and (3) even though the primary source of knowledge is one or more domain experts, user involvement in the development process remains an important factor for ES to have a desirable impact on the latter's jobs.     

基于单根I/O虚拟化的多根I/O资源池化方法

虚拟化技术在为现代数据中心提供高效的服务器整合能力和灵活的应用部署能力的同时,也对数据中心服务器的I/O系统设计提出了新的需求,现有I/O资源与服务器紧密绑定的I/O体系架构将产生成本上升、资源冗余、I/O连线复杂化等一系列问题.针对上述问题,提出了一种基于单根I/O虚拟化协议(single root I/O virtualization, SR-IOV)的多根I/O资源池化方法：基于硬件的多根域间地址和ID映射机制,实现了多个物理服务器对同一I/O设备的共享复用,有效减少单体服务器所需的设备数量和连线数量,并进一步提高服务器密度;同时提出虚拟I/O设备热插拔技术和多根共享管理机制,实现了虚拟I/O资源在服务器间的实时动态分配,提高资源的利用效率.提出的方法在可编程逻辑器件(fieid-programmable gate array, FPGA)原型系统中进行了验证,其评测表明,方法能够在实现多根I/O虚拟化共享的同时,保证各个根节点服务器获得近乎本地直连设备的I/O性能.     

Parallel fem algorithms based on recursive spatial decomposition—I. Automatic mesh generation

This paper discusses an automatic meshing scheme that is suitable for parallel processing. Meshes derived from solid models through recursive spatial decompositions inherit the hierarchical organization and the spatial addressability of the underlying grid. These two properties are exploited to design a meshing algorithm capable of operating in parallel (concurrent) processing environments. The concept of a meshing operator for parallel processing is defined and algorithms for various stages of the automatic meshing scheme are presented. A systematic simulation of fine- and coarse-grain parallel configurations is used to evaluate the performance of the meshing scheme. A companion paper focuses on parallel processing for the analysis of these automatically derived meshes via hierarchical substructuring.     

网络存储环境下的I／O负载

网络存储改变主机系统与存储设备间的连接关系，利用存储设备的计算能力直接向网络用户提供存储服务。这种结构的改变使系统存储处理模式相应发生变化，导致存储设备 I／O负载特征发生变化。本文提出存储系统负载不仅与应用环境相关，还与系统存储处理模式相关的观点，并据此分析了传统存储系统和网络存储系统的I／O负载特征。     

Matlab的数据文件I/O的实现方法

在实现Matlab和其它应用程序的数据共享时，Matlab常需要读写数据文件。本文详细讨论了Matlab数据文件I／O的实现方法，并给出了程序实例。     

PCMCIA接口及其接口卡设计

PCMCIA接口的物理特性和电气特性,并以PCMCIA接口“桥”电路芯片PCM16C010为例,结合小型单片机应用系统,阐述了接口卡的设计思路和关键技术,且给出了PCMCIA接口下I/O卡的设计实例.     

The impact of neural networks in finance

The financial industry is becoming more and more dependent on advanced computer technologies in order to maintain competitiveness in a global economy. Neural networks represent an exciting technology with a wide scope for potential applications, ranging from routine credit assessment operations to driving of large scale portfolio management strategies. Some of these applications have already resulted in dramatic increases in productivity. This paper brings together, from diverse sources, a collection of current research issues on neural networks in the financial domain. It examines a range of neural network systems related to financial applications from different levels of maturity to fielded products. It discusses the success rate of the neural network systems, and their performance in resolving particular financial problems.     

The impact of randomization in smoothing networks

We revisit randomized smoothing networks (Herlihy and Tirthapura in J Parallel Distrib Comput 66(5):626–632, 2006), which are made up of balancers and wires. We assume that balancers are oriented independently and uniformly at random. Tokens arrive arbitrarily on w input wires and propagate asynchronously through the network; each token is served on the output wire it arrives at. The smoothness is the maximum discrepancy among the numbers of tokens arriving at the w output wires. We present a collection of lower and upper bounds on smoothness, which are to some extent surprising: (1) The smoothness of a single block network is lg lg w+Θ(1) (with high probability), where the additive constant is between ?2 and 3. This tight bound improves vastly over the upper bound of ${\mathcal{O}(\sqrt{lg\, w})}$ from Herlihy and Tirthapura (2006), and it essentially settles our understanding of the smoothing properties of the block network. Further, this lg lg w+Θ(1) bound cannot be extended to universal smoothing a property stronger than smoothing that we introduce. (2) Most significantly, the smoothness of the cascade of two block networks is no more than 17 (with high probability); this is the first known randomized network with so small depth (2 lg w) and so good (constant) smoothness. The proof introduces some combinatorial and probabilistic structures and techniques which may be further applicable; the result demonstrates the full power of randomization in smoothing networks. (3) There is no randomized 1-smoothing network of width w and depth d that achieves 1-smoothness with probability better than ${\frac{\mathsf{d}} {w-1} }$ . In view of the deterministic, Θ(lg w)-depth, 1-smoothing network in Klugerman and Plaxton (Proceedings of the 24th Annual ACM Symposium on Theory of Computing, 1992), this result implies the first separation between deterministic and randomized smoothing networks. These results demonstrate an unexpected limitation on the power of randomization for smoothing networks: although it yields constant smoothness using small depth, going down to smoothness of 1 requires linear depth.     

The design and evaluation of fair scheduling in wireless mesh networks

In this paper we address the problem of scheduling in wireless mesh networks. First, we provide a comparison of existing scheduling algorithms and classify them based on the degree of fairness, the scheduling techniques and their implementation frameworks. Then we propose a fair scheduling approach using multiple gateways. The proposed scheduling approach consists of four important steps, namely, requirement tables, requirement propagation, clique generation and schedule generation. Simulation experiments are conducted to compare the performance of fair scheduling with the method that does not use fair scheduling. The simulation results confirm that the proposed scheduling has better performance with respect to the metrics used for performance evaluation.     

The I/O subsystem-a candidate for improvement

A computer system can be partitioned into hardware and the software executing on that hardware. The hardware consists of processor(s), memory, and “everything else”. The “everything else” we generally combine under the umbrella “I/O, whose job it is to manage the availability of information to and from the processor(s) and memory”. That information comes from storage devices, networks, and nonstorage devices. The I/O subsystem is the collection of all three; its influence on performance is a reflection of how well it manages the availability of information to and from all three. The impression today, from both the hardware side and the software side, is that the I/O subsystem can certainly stand improvement. The author considers improvements to the I/O subsystem     

并行I/O技术研究

从分析提高I／O性能的途径开始，对在分布主存的高性能计算机中利用存储系统并行性来完成数据访问的并行文件系统所涉及到的问题进行了分析和探讨，最后介绍了几个著名的并行文件系统。     

关于I/O模拟的讨论

详细分析Windows的I/O机制,提出了分别在用户级、系统级、驱动级的FO模拟操作,通过实际案例和源代码分享在I/O模拟上的得失,着重介绍鼠标与键盘的模拟.     

MPP系统中I/O数据直传的实现技术

传统MPP系统中ION对CN的读盘请求服务是先将数据从外设读到本地主存,再通过路由器将数据从本地主存发送到CN,此方式传输时延大,提出了ION中设备与CN间数据直传的方法,服务读盘请求时,直传技术不需经主存在的存储转发,因而能减少服务时延和提高I/O效率。     

面向随机读的磁盘阵列中的串内并发I／O技术研究

本文将SCSI总线协议中的失连一再选技术应用于磁盘阵列，使得磁盘阵列中同一串上各个磁盘之间高度并行，从而显著地提高随机读的性能。测试表明了并发技术的有效性。     

The role of spatial and typographic cues in the layout of journal references

Fifteen American and 15 British readers used the method of paired comparisons to assess 12 different ways of presenting reference lists in social science journals. The results indicated - despite British preferences for British styles - that the spatial arrangements of elements in the list was the major determinant of preference, and that the presence of a typographic cue only enhanced a preference for a particular layout.     

基于SYM53C8XX PCI-SCSI I/O处理器的多线程I/O调度

本文探讨了基于ＳＹＭ５３Ｃ８ＸＸ ＰＣＩ－ＳＣＳＩ　Ｉ／Ｏ处理器的多线程 Ｉ／Ｏ调度的算法及实现。并分析了一个多线程Ｉ／Ｏ调度的实例。