Application-controlled memory management in a single address space environment

With reference to a memory management system supporting the single address space abstraction and a uniform, persistent view of storage, we present a set of mechanisms that allow applications to exert explicit control over memory management activities. These mechanisms make it possible to move the contents of a virtual page to primary memory for fast processor access, or to push these contents back to secondary memory to free primary memory space. Our memory management scheme allows programs to exploit the memory reference pattern of the underlying algorithms, thereby improving utilisation of the system storage resources. This result is illustrated by using significant examples of memory management activities implemented at the application program level. Published online: 8 February 2001     

分布式状态空间生成的设计与实现

状态空间生成的并行化是针对状态空间爆炸问题而提出的一种重要手段。提出了一种基于MapReduce的分布式状态空间生成方案,与现有的同类研究相比,它无需用户关心生成算法的并行化,具有简单易用性;与常规的MapReduce的用法相比,它增加了输入文件的自动生成和作业运行的自动循环控制。该方案已在小规模分布式环境下实现,实验结果表明：（1）基于Map-Reduce的分布式状态空间生成算法可以扩大模型的可求解规模;（2）对于状态空间规模的增长主要由托肯（token）数增加引起的一类模型,该算法具有良好的适应性和可扩展性。     

从多地址空间到单地址空间再到无地址空间

在分析单地址空间操作系统的优点及其所存在问题的基础上,提出了无地址空间操作系统的思想,并介绍了一个无地址空间操作系统原型.在无地址空间操作系统中,没有进程虚拟空间的概念,指令直接对文件寻址,进程直接在文件上运行.与单地址空间操作系统相比,无地址空间操作系统不仅具有单地址空间操作系统的优点,而且能够避免单地址空间操作系统中存在的不足.     

空间高效的分布式数据存储方案

针对分布式数据存储中空间效率低、计算复杂度高等问题,基于 Jordan 矩阵和拉格朗日差值公式,提出了一种一般访问结构上高效的分布式数据存储方案。方案是计算安全的,空间利用率与理论安全的方案相比提高了 m2倍,每个存储服务器只需维护长度很短的秘密份额,就可以实现大数据的分布式存储。在数据存储过程中,存储服务器根据双线性对的性质计算并贡献影子份额,确保秘密份额的安全性。方案具有可公开验证性,有效防止了数据分发者与存储服务器的欺骗。最后对方案的正确性、安全性、拓展性、空间效率等进行分析,表明方案在分布式数据安全存储中具有很好的应用前景。     

分布式I/O日志回放系统的设计与实现

随着存储系统的快速发展,需要对不同存储系统的性能进行评测,以I/O日志为基础在实际应用环境中测试存储系统性能,更为客观和准确。提出了一种分布式环境下日志回放系统,通过中央控制器,能够方便地控制多个节点,同时对分布式存储系统进行性能测试,并对该系统的设计和实现过程进行了详细述。     

小型分布式控制系统中共享存储器的使用

介绍在小型分布式控制系统中,使用共享存储器作为主机和通信处理器的接口,以提高系统的实时响应能力。叙述了主机和通信处理器在访问共享存储器时可能发生的冲突的解决方法,以及主机和通信处理器接口协议及其实现方法。     

Exact controllability for singular distributed parameter system in Hilbert space

Exact controllability of singular distributed parameter control system is discussed via functional analysis and the theory of generalized operator semi-group in Hilbert space. Necessary and sufficient conditions concerning the exact controllability are given. Relations between exact controllability and stability of singular distributed parameter system are specified.     

面向新型非易失存储的文件系统研究综述

新型非易失存储(NVM)可字节寻址,具有近似内存的低延迟特性以及外存的非易失性,受限于软硬件技术成熟度,目前首先被用于外存.讨论了NVM用于持久性外存所面临的一系列问题,以及管理上的一些挑战;对现有的典型NVM文件系统及其主要特性进行了梳理.归纳起来,这些特性主要围绕降低一致性开销、降低软件栈开销、内存与外存的融合、分布式文件系统、NVM文件系统安全、容错、空间管理几个方面展开.最后,展望了NVM文件系统仍然有待探讨的几个研究方向,包括扩展性问题、虚拟内存与文件系统的有机融合以及分布式文件系统等.     

XpressSpace: a programming framework for coupling partitioned global address space simulation codes

Complex coupled multiphysics simulations are playing increasingly important roles in scientific and engineering applications such as fusion, combustion, and climate modeling. At the same time, extreme scales, increased levels of concurrency, and the advent of multicores are making programming of high‐end parallel computing systems on which these simulations run challenging. Although partitioned global address space (PGAS) languages attempt to address the problem by providing a shared memory abstraction for parallel processes within a single program, the PGAS model does not easily support data coupling across multiple heterogeneous programs, which is necessary for coupled multiphysics simulations. This paper explores how multiphysics‐coupled simulations can be supported by the PGAS programming model. Specifically, in this paper, we present the design and implementation of the XpressSpace programming system, which extends existing PGAS data sharing and data access models with a semantically specialized shared data space abstraction to enable data coupling across multiple independent PGAS executables. XpressSpace supports a global‐view style programming interface that is consistent with the PGAS memory model, and provides an efficient runtime system that can dynamically capture the data decomposition of global‐view data‐structures such as arrays, and enable fast exchange of these distributed data‐structures between coupled applications. In this paper, we also evaluate the performance and scalability of a prototype implementation of XpressSpace by using different coupling patterns extracted from real world multiphysics simulation scenarios, on the Jaguar Cray XT5 system at Oak Ridge National Laboratory. Copyright © 2013 John Wiley & Sons, Ltd.     

Treating data privacy in distributed systems

This papers describes and is primarily concerned with the security data definition and management in a distributed data base of aggregated type, although the approach described may be applied to any distributed system architecture. A multi-level logical security architecture is presented reflecting the logical architecture of the distributed system. In particular, three security logical schemata are proposed: the network security schema, the external security schemata, and the intermediate security schemata. For each schema data models are introduced, allowing the definition and the management of security information. Mapping rules between the logical levels are discussed. Finally security mechanisms are analyzed.     

基于分布式中间件的电力监控系统设计与实现

基于对象的分布式计算技术是当今计算机软件开发的所采用的一种重要技术.与传统开发技术相比,分布式对象技术具有更好的开放性和扩展性.基于对象的中间件是分布式对象技术的良好应用.结合能量管理系统的需要,在分布式对象互操作思想的基础上,对电力监控系统中所采用的分布式中间件技术进行了设计与分析,对电力监控系统中的中间件技术的组织模式、系统流程及实现模式进行了设计,应用cORBA中间件技术对其进行了实现,并在某智能配电自动化有限公司得到了相应的应用.     

Object and process migration in a single-address-space distributed system

In a distributed system framework, program interactions can be modelled by using typed objects according to client/server relationships. The operations defined by a given type are the services that may be provided by an object of this type to a client process. When the process and the object are located in different nodes, migrations may represent valid alternatives to remote procedure calls. Migration of the server object causes the memory area storing the internal representation of this object to be copied into the node of the client process. Migration of the client process causes execution of this process to proceed in the node of the server object. This paper proposes migration paradigms with reference to a memory environment implementing the notion of a single address space. The discussion takes a number of salient issues into consideration, including performance, memory configurations for object storage, and the strategies for memory management.     

Work stealing for GPU‐accelerated parallel programs in a global address space framework

Task parallelism is an attractive approach to automatically load balance the computation in a parallel system and adapt to dynamism exhibited by parallel systems. Exploiting task parallelism through work stealing has been extensively studied in shared and distributed‐memory contexts. In this paper, we study the design of a system that uses work stealing for dynamic load balancing of task‐parallel programs executed on hybrid distributed‐memory CPU‐graphics processing unit (GPU) systems in a global‐address space framework. We take into account the unique nature of the accelerator model employed by GPUs, the significant performance difference between GPU and CPU execution as a function of problem size, and the distinct CPU and GPU memory domains. We consider various alternatives in designing a distributed work stealing algorithm for CPU‐GPU systems, while taking into account the impact of task distribution and data movement overheads. These strategies are evaluated using microbenchmarks that capture various execution configurations as well as the state‐of‐the‐art CCSD(T) application module from the computational chemistry domain. Copyright © 2016 John Wiley & Sons, Ltd.     

一种分布式系统中线程池的设计与实现

随着网络和分布式计算的日益发展,负载数量迅速增长,如何在大负载情况下保证高效的吞吐率是构建服务器时所面临的问题,目前广泛采用的技术之一是线程池。借鉴Leader／Followers模式,设计并实现了一类高效的线程池,并针对分布式系统环境做了进一步的改进。此方法已经应用到分布事务监控器OnceTX的实现中,并取得了很好的效果。     

分布式存储系统中的低修复成本纠删码

纠删码技术是分布式存储系统中典型的数据容错方法，与多副本技术相比，能够以较低的存储开销提供较高的数据可靠性；然而，纠删码修复成本过高的特点限制了其应用。针对现有纠删码修复成本高、编码复杂和灵活性差的问题，提出一种编码简单的低修复成本的纠删码——旋转分组修复码（RGRC）。RGRC首先将多个条带组合成条带集，然后利用条带之间的关联关系对条带集内的数据块进行分层旋转编码，以此得到相应的冗余块。RGRC大幅度地减少了单节点修复过程中所需要读取和传输的数据量，从而能节省大量的网络带宽资源。同时RGRC在解决单节点修复成本高的问题时，依然保留着较高的容错能力，且为满足分布式存储系统的不同需求，可以灵活地权衡系统的存储开销和修复成本。在分布式存储系统中进行的对比实验分析结果展示，与其他常用的RS（Reed-Solomon）码、LRC（Locally Repairable Codes）、basic-Pyramid、DLRC（Dynamic Local Reconstruction Codes）、pLRC（proactive Locally Repairable Codes）、GRC（Group Repairable Codes）、UFP-LRC（Unequal Failure Protection based Local Reconstruction Codes）相比，RGRC只需要增加少量的存储开销，就能降低单节点修复14%~61%的修复成本，同时减少14%~58%的修复时间。     

存储系统的层次性与进程数据存储模型

计算机系统的各种存储部件,包括寄存器、cache、内存、外存,被组织成一个层次性的存储结构,传统操作系统分别在内存和外存这两个层次上为用户构造了两种数据存储模型：进程逻辑空间和文件。应该在存储层次最外层--外存直接为用户构造进程数据存储模型--文件,把全部内存作为进程访问外存的缓冲,取消了进程逻辑空间的概念,实现指令对文件的直接寻址,使进程直接在文件上运行,基于以上思想,提出了一个操作系统模型,介绍了该模型的结构及其一个实现原形,并对其性能进行了测试和评价,最后讨论了该操作系统具有的优点及其存在的问题。     

A Fault-tolerant model for tuple space coordination in distributed environments

In distributed systems, tuple space is one of the coordination models that significantly maximizes system performance against failure due to its space and time decoupling features. With the growing popularity of distributed computing and increasing complexity in the network, host and link failure occurs frequently, resulting in poor system performance. This article proposes a fault-tolerant model named Tuple Space Replication (TSR) for tuple space coordination in distributed environments. The model introduces a multi-agent system that consists of multiple hosts. Each host in a multi-agent system comprises an agent space with a tuple space for coordination. In this model, we introduce three novel fault-tolerant algorithms for tuple space primitives to provide coordination among hosts with tolerance to multiple links and hosts failure. The first algorithm is given for out() operation to insert tuples in the tuple space. The second algorithm is presented for rdp() operation to read any tuple from the tuple space. The third algorithm is given for inp() operation to delete or withdraw tuples from the tuple space. These algorithms use less number of messages to ensure consistency in the system. The message complexity of the proposed algorithms is analyzed and found O(n) for out(), O(1) for rdp(), and O(n) for inp() operations which is comparable and better than existing works, where n is the number of hosts. The testbed experiment reveals that the proposed TSR model gives performance improvement up to 88%, 70.94%, and 63.80% for out(), rdp(), and inp() operations compared to existing models such as FT-SHE, LBTS, DEPSPACE, and E-DEPSPACE.     

分布式存储系统中的低修复成本纠删码

纠删码技术是分布式存储系统中典型的数据容错方法,与多副本技术相比,能够以较低的存储开销提供较高的数据可靠性;然而,纠删码修复成本过高的特点限制了其应用。针对现有纠删码修复成本高、编码复杂和灵活性差的问题,提出一种编码简单的低修复成本的纠删码——旋转分组修复码（RGRC）。RGRC首先将多个条带组合成条带集,然后利用条带之间的关联关系对条带集内的数据块进行分层旋转编码,以此得到相应的冗余块。RGRC大幅度地减少了单节点修复过程中所需要读取和传输的数据量,从而能节省大量的网络带宽资源。同时RGRC在解决单节点修复成本高的问题时,依然保留着较高的容错能力,且为满足分布式存储系统的不同需求,可以灵活地权衡系统的存储开销和修复成本。在分布式存储系统中进行的对比实验分析结果展示,与其他常用的RS（Reed-Solomon）码、LRC（Locally Repairable Codes）、basic-Pyramid、DLRC（Dynamic Local Reconstruction Codes）、pLRC（proactive Locally Repairable Codes）、GRC（Group Repairable Codes）、UFP-LRC（Unequal Failure Protection based Local Reconstruction Codes）相比,RGRC只需要增加少量的存储开销,就能降低单节点修复14%~61%的修复成本,同时减少14%~58%的修复时间。     

分布式信息系统集成的语义架构SP4IDIS

为了解决分布式信息系统之间信息无法共享的问题,提出一种基于形式语义的系统集成支撑技术,构建了分布式信息系统的可共享语义平台.该方法保持了源信息系统的自治性,通过构造融合本体为分布式信息系统提供信息解释公用模版.与已有的方法相比,该系统采全形式化语义支撑,奠定了分布式信息系统自动化集成的基础.