面向异构多核处理器的并行代价模型

现有的并行代价模型大多是面向共享存储或分布存储结构设计的,不完全适合异构多核处理器。为解决这个问题,提出了面向异构多核处理器的并行代价模型,通过定量刻画计算核心运算能力、存储访问延迟和数据传输开销对循环并行执行时间的影响,提高加速并行循环识别的准确性。实验结果表明,提出的并行代价模型能有效识别加速并行循环,将其识别结果作为后端生成并行代码的依据,可有效提高并行程序在异构多核处理器上的性能。     

基于OMNet++的大规模InfiniBand互连网络模拟系统

随着多核处理器的发展和计算需求的不断增长,高性能计算系统规模不断增大.使用模拟器对高性能计算系统进行模拟,对系统设计及优化有着重要的作用,互连网络模拟则是其中不可或缺的一部分.设计实现了一种基于OM Net++的大规模InfiniBand互连网络模拟系统,该系统通过记录的并行程序M PI消息来驱动网络仿真过程,可以模拟...     

A Heterogeneous Mixed-Mode Execution Model for Massively Parallel Systems

In this paper, we consider a massively parallel system that is composed of heterogeneous processors, that is, processors with different processing power, and that combines the advantages of the SIMD and MIMD architectures. The heterogeneous mixed-mode (HeMM) execution model is composed of two main components, which operate in the well-known SIMD and MIMD paradigms. The main computing power comes from a component that is composed of a massive number of processors and operates in a data parallel manner. The other component is composed of a few (or even one) fast processors which operate in the MIMD paradigm. The operation of a small number of processors in an MIMD paradigm has been well demonstrated through actual systems. The processors in this component add flexibility to the execution of the parallel programs such that it adjusts to the changing parallelism of the program to enhance the performance. Based on this execution model we analyze the gains in performance that is obtainable by this new system. We show that substantial performance gains can be obtained by using the HeMM system.     

Web services based scheduling in OpenCF

Cloud Computing is emerging as a new computing paradigm which aims to provide reliable, customized and QoS guaranteed dynamic computing environments for end users. The availability of these large, virtualized pools of computing resources raises the possibility of a new computing paradigm for scientific research with many advantages. For research groups, Cloud Computing provides convenient access to reliable, high-performance clusters and storage without the need to purchase and maintain sophisticated hardware. For developers, virtualization allows scientific codes to be optimized and pre-installed on machine images, facilitating control over the computational environment. In these large-scale, heterogeneous and dynamic systems, the efficient execution of parallel computations can require mappings of tasks to computing resources whose performance is both irregular (because of heterogeneity) and variable in time (because of dynamicity). This paper introduces our initial experience with Cloud Computing based on a Python implementation of our OpenCF framework. We propose to show the features provided by OpenCF using the Google Application Engine as a proof of concept.     

基于事务性执行的投机并行多线程软件模拟

基于事务性执行的投机并行多线程是一种适合未来多核微处理器架构的新型并行程序设计和编译技术.但在此基础上的并行程序执行过程更为复杂,程序执行过程的模拟成为关键问题之一.本文提出利用二进制代码级动态插桩技术对投机并行多线程程序进行功能性模拟,设计并实现了完整的软件平台,可精确地模拟和监控并行程序的线程级投机执行过程,检测访存冲突,从而实现投机并行多线程的语义.该软件平台同时可以作为进一步研究投机多线程并行程序真实执行过程的基础,并有效支持投机并行多线程编译器的设计和分析.     

A novel strategy for building interoperable MPI environment in heterogeneous high performance systems

Breakthrough advances in microprocessor technology and efficient power management have altered the course of development of processors with the emergence of multi-core processor technology, in order to bring higher level of processing. The utilization of many-core technology has boosted computing power provided by cluster of workstations or SMPs, providing large computational power at an affordable cost using solely commodity components. Different implementations of message-passing libraries and system softwares (including Operating Systems) are installed in such cluster and multi-cluster computing systems. In order to guarantee correct execution of message-passing parallel applications in a computing environment other than that originally the parallel application was developed, review of the application code is needed. In this paper, a hybrid communication interfacing strategy is proposed, to execute a parallel application in a group of computing nodes belonging to different clusters or multi-clusters (computing systems may be running different operating systems and MPI implementations), interconnected with public or private IP addresses, and responding interchangeably to user execution requests. Experimental results demonstrate the feasibility of this proposed strategy and its effectiveness, through the execution of benchmarking parallel applications.     

Core frequency adjustment to optimize Time Warp on many-core processors

Time Warp synchronized parallel discrete event simulators are organized to operate asynchronously and aggressively without explicit synchronization between the concurrently executing simulation objects. In place of an explicit synchronization mechanism, the concurrent simulators implement an independent but common virtual clock model and a rollback/recovery mechanism to restore causal order when out-of-order events are detected. When the critical path of execution of the simulation is balanced across this parallel threads of execution, this can result in a highly effective, lightweight synchronization mechanism to implement parallel simulation. However, imbalances in the workload across the threads can result in excessive rollback in some threads and slowed progress of the critical path. On small shared memory multi-core systems, a lowest time-stamp scheduling policy can effectively balance the workload. However, on larger many-core chips, conventional load balancing and workload migration will once again become necessary. Fortunately, emerging many-core chips contain some interesting features that can potentially be exploited to improve the performance of parallel simulations. In particular, the recently developed Intel Single-chip Cloud Computer (SCC) provides mechanisms for the runtime control of the frequency and voltage settings of the chip. Furthermore, the frequency and voltage settings are independently set within different regions (called islands) of the chip. Thus, in a Time Warp simulation, one could increase the frequency of the cores executing threads on the critical path (those experiencing infrequent rollback) and decrease the frequency of the cores executing threads off the critical path (those experiencing excessive rollback). This paper investigates the run-time control and adjustment of core frequency in some contemporary x86 multi-core processors to identify the platforms that can support the exploration of dynamic run-time control of core frequency settings. The results show that while all multi-core processors have software controllable core frequency modulation capabilities, they are generally not fully independent as the system comes under load and are therefore unsuitable for these studies. Fortunately, one processor, the AMD X6 line, provides software control for core frequencies that can be fixed (by software) even as the system operates under load.     

A high performance framework for modeling and simulation of large-scale complex systems

Due to the quick advances in the scale of problem domain of complex systems under investigation, the complexity of multi-input component models used to construct logical processes (LP) has significantly increased. High-performance computing technologies have therefore been extensively used to enable parallel simulation execution. However, the traditional multi-process parallel method (MPM) executes LPs in parallel on multi-core platforms, which ignores the intrinsic parallel capabilities of multi-input component models. In this study, a vectorized component model (VCM) framework has been proposed. The design aims to better utilize the parallelism of multi-input component models. A two-level composite parallel method (CPM) has then been constructed within the framework, which can sustain complex system simulation applications consisting of multi-input component models. CPM first employs MPM to dispatch LPs onto a multi-core computing platform. It then maps VCMs to the multiple-core platform for parallel execution. Experimental results indicate that (1) the proposed VCM framework can better utilize the parallelism of multi-input component models, and (2) CPM can significantly improve the performance comparing to the traditional MPM. The results also show that CPM can effectively cope with the size and complexity of complex simulation applications with multi-input component models.     

异构多核处理器体系结构设计研究

多核技术成为当今处理器发展的重要方向,异构多核处理器由于可将不同类型的计算任务分配到不同类型的处理器核上并行处理,从而为不同需求的应用提供更加灵活、高效的处理机制而成为当今研究的热点.本文从体系结构的角度探讨了异构多核处理器设计中的关键点,从内核结构、互连方式、存储系统、操作系统支持、测试与验证、动态电压调节等方面分析...     

A software framework for fine grain parallelization of cellular models with OpenMP: Application to fire spread

We are dealing here with the parallelization of fire spreading simulations following detailed physical experiments. The proposal presented in this paper has been tested and evaluated in collaboration with physicists to meet their requirements in terms of both performance and precision. For this purpose, an object-oriented framework using two abstraction levels has been developed. A first level considers the simulation as a global phenomenon which evolves in space and time. A local level describes the phenomena occurring on elementary parts of the domain. In order to develop an extensible and modular architecture, the cellular automata paradigm, the DEVS discrete event system formalism and design patterns have been used. Simulation treatments are limited to a set of active elements to improve execution times. A new kind of model, called Active-DEVS is then specified. The model is computed with a fine grain parallelization very efficient for present day multi-core processors which are elementary units of modern computing clusters and computing grids. In this paper, the parallelization with Open MultiProcessing (OpenMP) standard directives on Symmetric MultiProcessing (SMP) architectures is discussed and the efficiency of the retained solution is studied.     

FASA: A software architecture and runtime framework for flexible distributed automation systems

Modern automation systems have to cope with large amounts of sensor data to be processed, stricter security requirements, heterogeneous hardware, and an increasing need for flexibility. The challenges for tomorrow’s automation systems need software architectures of today’s real-time controllers to evolve.This article presents FASA, a modern software architecture for next-generation automation systems. FASA provides concepts for scalable, flexible, and platform-independent real-time execution frameworks, which also provide advanced features such as software-based fault tolerance and high degrees of isolation and security. We show that FASA caters for robust execution of time-critical applications even in parallel execution environments such as multi-core processors.We present a reference implementation of FASA that controls a magnetic levitation device. This device is sensitive to any disturbance in its real-time control and thus, provides a suitable validation scenario. Our results show that FASA can sustain its advanced features even in high-speed control scenarios at 1 kHz.     

Parallel computation of symbolic robot models and control laws: Theory and application to transputer networks

New computer architectures based on large numbers of processors are now used in various application areas ranging from embedded systems to supercomputers. Efficient parallel processing algorithms are applied in a wide variety of applications such as simulation, robot control, and image synthesis. This article presents two novel parallel algorithms for computing robot inverse dynamics (as well as control laws) starting from customized symbolic robot models. To gain the most benefit from the concurrent processor architecture, the whole job is divided into a large number of simple tasks, each involving only a single floating-point operation. Although requiring sophisticated scheduling schemes, fine granularity of tasks was the key factor for achieving nearly maximum efficiency and speedup. The first algorithm resolves the scheduling problem for an array of pipelined processors. The second one is devoted to parallel processors connected by a complete crossbar interconnection network. The main feature of the proposed algorithms is that they take into account the communication delays between processors and minimize both the execution time and communication cost. To prove the theoretical results, the algorithms have been verified by experiments on an INMOS T800 transputer-based system. We used four transputers in serial and parallel configurations. The experimental results show that the most complicated dynamic control laws can be executed in a submilisecond time interval. © 1993 John Wiley & Sons, Inc.     

Delivering cloud services with QoS requirements: Business opportunities,architectural solutions and energy-saving aspects

The flexible and pay-as-you-go computing capabilities offered by Cloud infrastructures are nowadays very attractive, and widely adopted by many organizations and enterprises. In particular this is true for those having periodical or variable tasks to execute, and, choose to not or cannot afford the expenses of buying and managing computing facilities or software packages, that should remain underutilized for most of the time. For their ability to couple the scalability offered by public service providers, with the wider Quality of Service (QoS) provisions and ad-hoc customizations provided by private Clouds, Hybrid Clouds (HC) seem a particularly appealing solution for customers requiring something more than the mere availability of the service.The paper firstly introduces a Cloud brokering system leveraging on a promising architectural approach, based on the use of a gateway toolkit. This approach provides noticeably advantages, both to customers and to Cloud Brokers (CB), for its ability to hide all the intricacies related to the management of powerful, but often complex and heterogeneous infrastructures like the Cloud. Moreover such approach, through customized interfaces, facilitates customers in accessing Cloud resources thus easing the tailored deployment and execution of their applications and workflows.The major contribution of this work is given by the analysis of a set of brokering strategies for Hybrid Clouds, implemented by a brokering algorithm, aimed at the execution of various applications subject to different user requirements and computational conditions. With the objective of firstly maximize both user satisfaction and CB’s revenues the algorithm also pursues profit increases through the reduction of energy costs by adopting energy saving mechanisms. A simulation model is used to evaluate performance, and the results show that differences among strategies depend on type and size of system loads and that the use of turn on and off techniques greatly improves energy savings at low and medium load rates thus indirectly increasing CB revenues without diminishing customers’ satisfaction.     

PHR: A Parallel Hierarchical Radiosity System with Dynamic Load Balancing

In this paper, we present a parallel system called PHR for computing hierarchical radiosity solutions of complex scenes. The system is targeted for multi-processor architectures with distributed memory. The system evaluates and subdivides the interactions level by level in a breadth first fashion, and the interactions are redistributed at the end of each level to keep load balanced. In order to allow interactions freely travel across processors, all the patch data is replicated on all the processors. Hence, the system favors load balancing at the expense of increased communication volume. However, the results show that the overhead of communication is negligible compared with total execution time. We obtained a speed-up of 25 for 32 processors in our test scenes.     

Characterization of server performance bottlenecks in distributed interactive simulation environments

Servers running distributed simulation applications need to process a large number of small packets at the network level. This becomes a burden on the server CPUs and exhibits itself as high CPU utilization. In this paper, first, we confirm this phenomenon, then, we utilize performance analyzer tools from major processor vendors to characterize the major sources of CPU usage. In addition, the experiments run on real hardware reveal server bottlenecks within the context of new architectural features such as hyper-threading, hyper-transport and multi-core processors.     

基于多核阵列体系结构的嵌套循环并行优化

多核处理器已广泛应用于高性能计算领域,如何有效地将传统串行程序转换为并行代码并减少程序中嵌套循环所占用时间仍是该领域的挑战性问题。本文首先基于多面体模型对嵌套循环进行依赖特征分析并实现瓦片分割,据此自动生成粗粒度并行代码。针对多核阵列处理器的结构特点,采用遗传算法生成通信优化的瓦片任务序列,在此基础上建立了有效的任务调度模型。最后将上述方法应用于LU分解,结果表明该方法与传统调度算法相比,在增加数据局部性、实现负载平衡方面具有更好效果。     

Balancing throughput and response time in online scientific Clouds via Ant Colony Optimization (SP2013/2013/00006)

The Cloud Computing paradigm focuses on the provisioning of reliable and scalable infrastructures (Clouds) delivering execution and storage services. The paradigm, with its promise of virtually infinite resources, seems to suit well in solving resource greedy scientific computing problems. The goal of this work is to study private Clouds to execute scientific experiments coming from multiple users, i.e., our work focuses on the Infrastructure as a Service (IaaS) model where custom Virtual Machines (VM) are launched in appropriate hosts available in a Cloud. Then, correctly scheduling Cloud hosts is very important and it is necessary to develop efficient scheduling strategies to appropriately allocate VMs to physical resources. The job scheduling problem is however NP-complete, and therefore many heuristics have been developed. In this work, we describe and evaluate a Cloud scheduler based on Ant Colony Optimization (ACO). The main performance metrics to study are the number of serviced users by the Cloud and the total number of created VMs in online (non-batch) scheduling scenarios. Besides, the number of intra-Cloud network messages sent are evaluated. Simulated experiments performed using CloudSim and job data from real scientific problems show that our scheduler succeeds in balancing the studied metrics compared to schedulers based on Random assignment and Genetic Algorithms.     

A modeling and simulation framework for mobile cloud computing

Mobile cloud computing (MCC) is an emerging paradigm for transparent elastic augmentation of mobile devices capabilities, exploiting ubiquitous wireless access to cloud storage and computing resources. MCC aims at increasing the range of resource-intensive tasks supported by mobile devices, while preserving and extending their resources. Its main concerns regard the augmentation of energy efficiency, storage capabilities, processing power and data safety, to improve the experience of mobile users. The design of MCC systems is a challenging task, because both the mobile device and the Cloud have to find energy-time tradeoffs and the choices on one side affect the performance of the other side. The analysis of the MCC literature points out that all existing models focus on mobile devices, considering the Cloud as a system with unlimited resources. Also, to the best of our knowledge, no MCC-specific simulation tool exists. To fill this gap, in this paper, we propose a modeling and simulation framework for the design and analysis of MCC systems, encompassing all their components. The main pillar of the proposed framework is the autonomic strategy consisting of adaptive loops between every mobile devices and the Cloud. The proposed model of the mobile device takes into account online estimations of the actual Cloud performance – not only the nominal values of the performance indicators. At the same time, the model of the Cloud takes into consideration the characteristics of the workload, to adapt its configuration in terms of active virtual machines and task management strategies. Moreover, the developed discrete event simulator is an effective tool for the evaluation of an MCC system as a whole, or single components, considering different classes of parallel jobs.     

Using Byzantine Fault-Tolerance to Improve Dependability in Federated Cloud Computing

Computing Clouds are typically characterized as large scale systems that exhibit dynamic behavior due to variance in workload. However, how exactly these characteristics affect the dependability of Cloud systems remains unclear. Furthermore provisioning reliable service within a Cloud federation, which involves the orchestration of multiple Clouds to provision service, remains an unsolved problem. This is especially true when considering the threat of Byzantine faults. Recently, the feasibility of Byzantine Fault-Tolerance within a single Cloud and federated Cloud environments has been debated. This paper investigates Cloud reliability and the applicability of Byzantine Fault-Tolerance in Cloud computing and introduces a Byzantine fault-tolerance framework that enables the deployment of applications across multiple Cloud administrations. An implementation of this framework has facilitated in-depth experiments producing results comparing the reliability of Cloud applications hosted in a federated Cloud to that of a single Cloud.