Distributed scheduling with probabilistic and fuzzy classifications of processes

The classification of various types of processes is an important factor in large-scale distributed systems such as, grid and cloud platforms. Moreover, the coordination and control of distributed processes are research challenges in presence of unpredictable network partitioning and distributed semaphores. The process classification is important in order to allocate and schedule distributed processes enhancing overall resource utilization and throughput. The schedulers employ patterns of resource affinities of concurrent processes in order to make scheduling decisions affecting overall resource utilization in a system, where resource affinity patterns of a process may not be static. This paper proposes an estimation model and a classifier algorithm to queuing processes based on respective resource affinities. The kernel-level software architecture is designed to control scheduling of distributed processes based on classification for enhanced throughput. The classifier algorithm tracks the resource affinities of processes based on execution traces and the control algorithm performs process scheduling. Experimental results indicate that the classifier algorithm successfully manages process queues based on resource affinities of processes and, the control algorithm successfully monitors scheduler activation for a set of processes.     

Symbian操作系统环境下通信系统的设计与实现

Symbian操作系统是一种32位抢占式多任务操作系统,具有功耗低、内存占用少等特点,非常适合手机等移动设备使用。它提供了广泛的通信基础结构,包括串行通信、TCP/IP、IrDA、USB、传真、以及通过GSM和GPRS的通信。本文介绍了如何利用Symbian操作系统提供的各种通信机制来构建复杂高效的通信程序。     

Software system of the Earth Simulator

The Earth Simulator (ES) is a large scale, distributed memory, parallel computer system consisting of 640 processor nodes (PN) with shared memory vector multi-processors (64GFLOPS/PN, 5120 APs in total, AP: arithmetic processor). All the nodes are connected via a high speed (16GB/s) single-stage crossbar network called the Interconnection Network (IN).

The operating system for the Earth Simulator is based on SUPER-UX, the UNIX operating system for the SX series scientific supercomputers. In order to realize high-performance parallel processing on the highly parallel machine, the operating system is enhanced for scalability.

The Earth Simulator system is managed as a two-level cluster system called the Super Cluster System. In the Super Cluster System, the Earth Simulator system is divided into 40 clusters (16 PNs/cluster). A single controller called Super Cluster Control Station (SCCS) manages all these clusters. This management system provides Single System Image (SSI) operation, management and job control for the large scale multi-node system.

The Job Scheduler (JS) and NQS running on the SCCS control all jobs of the system. They schedule the resources such as processing nodes and files which have not usually been treated as scheduling resources. This allows efficient scheduling of large scale jobs.

The MPI library (MPI/ES) and the HPF compiler (HPF/ES) are available for distributed parallel programming on the Earth Simulator. MPI/ES conforms to the MPI 2.0 standard and is optimized to exploit the hardware features. HPF/ES conforms to the core part of HPF 2.0 and supports some features of the HPF 2.0 approved extensions and HPF/JA 1.0 extensions. HPF/ES suitably handles the 3-level parallelism of the Earth Simulator system, that is, vectorization, shared-memory parallelization, and distributed-memory parallelization. Moreover, HPF/ES extends the language to easily handle irregular problems.     

嵌入式实时系统调度策略

嵌入式系统在当今的生产和生活中得到了广泛的应用,鉴于嵌入式实时系统的特点,要求任务调度等实时内核功能精简和高效,综合了EDF和RM调度策略的CSD调度策略,更加适合嵌入式系统的特点,满足其内核的要求。     

基于NPU的IP-DSLAM中包调度器的研究

为了满足IP-DSLAM中包调度器高速处理和QoS保证的设计目标,采用分层调度的思想,结合Intel IXP2350网络处理器多核多线程的体系结构特征,研究一种基于网络处理器的IP-DSLAM中HHRR包调度器的设计与实现.实验表明,HHRR调度器实现简单高效,实现了IP-DSLAM的线速转发,有利于保证不同业务类的服务质量要求.该项研究为网络应用中包调度器的设计,研究调度算法和QoS保证提供了依据.     

The implementation of IEEE 802.16m protocol module for ns-3 simulator

This paper presents the first IEEE 802.16m medium access control (MAC) protocol module embedded in the network simulator version 3 (ns-3). The designed module provides a validation tool for researchers to verify simulated results related to the IEEE 802.16m. This module supports the basic MAC functions which includes an initial ranging (IR) function; four types of bandwidth request (BR) procedures; five types of standard-specified quality-of-service (QoS); a QoS-based transmission scheduler; an adaptive modulation and coding (AMC) selection function; a physical (PHY) layer with a channel propagation model. The designed module provides modification flexibility to adapt to other simulation modules. Various scenarios are simulated to show the capacity of the implemented IEEE 802.16m module in details. Results of different types of traffic are tested with the path loss effect on transmission to visualize the numeric capability of the module.     

Scheduling with accurate communication delay model and scheduler implementation for multiprocessor system-on-chip

In multiprocessor system-on-chip, tasks and communications should be scheduled carefully since their execution order affects the performance of the entire system. When we implement an MPSoC according to the scheduling result, we may find that the scheduling result is not correct or timing constraints are not met unless it takes into account the delays of MPSoC architecture. The unexpected scheduling results are mainly caused from inaccurate communication delays and or runtime scheduler’s overhead. Due to the big complexity of scheduling problem, most previous work neglects the inter-processor communication, or just assumes a fixed delay proportional to the communication volume, without taking into consideration subtle effects like the communication congestion and synchronization delay, which may change dynamically throughout tasks execution. In this paper, we propose an accurate scheduling model of hardware/software communication architecture to improve timing accuracy by taking into account the effects of dynamic software synchronization and detailed hardware resource constraints such as communication congestion and buffer sharing. We also propose a method for runtime scheduler implementation and consider its performance overhead in scheduling. In particular, we introduce efficient hardware and software scheduler architectures. Furthermore, we address the issue of centralized implementation versus distributed implementation of the schedulers. We investigate the pros and cons of the two different scheduler implementations. Through experiments with significant demonstration examples, we show the effectiveness of the proposed approach.     

一种Hadoop中基于作业类别和截止时间的调度算法

Hadoop是一种开源可靠的分布式计算框架,而MapReduce是处理超大规模数据集的编程模型.鉴于Ha-doop内置的调度器不能很好地处理类别不同且有截止时间的作业的调度,提出了一种基于作业类别和截止时间的作业调度算法.作业分为CPU密集型和I/O密集型,并根据截止时间设置优先级来实现作业的调度.实验结果表明,该算法在充分利用集群的CPU和磁盘I/O的同时,能满足作业的截止期需求,当同一时间段内截止时间相近时算法达到最优,当某一队列中作业截止时间均比另一种队列短时,算法效率最低.     

实时嵌入式系统中的一种互斥方法

本文分析了实时嵌入式操作系统中几种常用的互斥方法,提出了一种基于任务优先级的禁止抢占法,并分析了该方法的安全性和高效性。     

适应于无记忆进程的pGVT算法和调度策略

TimeWarp并行离散事件模拟中,事件和状态信息被保留,用于可能的回滚处理,时间戳小于GVT的事件和状态不再被保存。针对无记忆进程不对中间状态产生影响的特点,在不影响其他进程执行的前提下,改进已有的pGVT算法和调度策略,减少相关的通讯量和可能出现的回滚,提高整个模拟的效率。