QoS in WiMAX hybrid schedulers for heterogeneous traffic and their performance comparison

ABSTRACT

Worldwide Interoperability for Microwave Access (WiMAX), based on IEEE 802.16, provides first-mile wireless access for broadband users and is capable of satisfying the Quality of Service (QoS) requirements of voice, video and data traffic (triple-play services). This paper proposes two WiMAX hybrid schedulers, developed by integration of homogeneous schedulers, for providing QoS to triple-play services. The Markovian model is solved analytically to derive various performance metrics, which are then compared with an existing scheduler. The splitting of FTP traffic, before scheduling, provides improved throughput as compared to the proposed first hybrid scheduler, making this an exclusive component of the proposed second hybrid scheduler. The results show that the second proposed model provides an improvement of mandatory QoS parameters; reduction in voice traffic mean queuing delay by 68.18%, and improvement in FTP traffic throughput by 67.27% as compared to the first designed model. The first considered scheduler gives high inter-class fairness; however, it does not provide satisfactory QoS performance. The proposed hybrid schedulers are better in overall performance than other schedulers proposed recently, as they not only satisfy the QoS demands of different types of services but also improve the fairness among services.     

Schedulability criterion and performance analysis of coordinated schedulers

Inter-server coordinated scheduling is a mechanism for downstream nodes to increase or decrease a packet's priority according to the congestion incurred at upstream nodes. In this paper, we derive an end-to-end schedulability condition for a broad class of coordinated schedulers that includes Core-stateless Jitter Virtual Clock (CJVC) and Coordinated Earliest Deadline First (CEDF). In contrast to previous approaches, our technique purposely allows flows to violate their local priority indexes while still providing an end-to-end delay bound. We show that under a simple priority assignment scheme, coordinated schedulers can outperform WFQ schedulers, while replacing per-flow scheduling operations with a simple coordination rule. Finally, we illustrate the performance advantages of coordination through numerical examples and simulation experiments.     

Hadoop HDFS和MapReduce架构浅析

在简述Hadoop namenode、datanode运行模式的基础上,重点介绍了Hadoop MapReduce的工作机制,并以作业提交、作业初始化、任务分配、任务执行和任务进度更新等流程介绍了Job Client、JobTracker、TaskTracker和HDFS在MapReduce过程中的分工与协作,最后,对云计算作出展望。     

面向大数据的数据服务中心设计与应用研究

大数据管理对传统的基于关系数据库管理系统为核心的数据管理服务平台提出了挑战,研究开发支持结构化和非结构化数据统一管理的数据管理与服务中心成为一项非常迫切并且重要的任务,汇总分析出大数据服务中心四项关键技术,采用可配置泛化表模型等技术,设计了基于Hadoop平台的数据服务中心系统原型BIG-DMS和基于BIG-DMS的数据综合服务平台应用方案,支持结构化和非结构化数据的统一存储、管理和访问服务。     

大数据开源技术发展研究

大数据技术发展与开源运动的结合已成为大数据技术创新中的一个鲜明特点。目前,大数据分析处理流程中所使用的关键技术几乎都源自开源模式,知名的大数据开源项目如分布式计算和存储系统Hadoop、基于内存计算的集群计算系统Spark,以及多款非关系型数据库（NoSQL）产品等。文章对Hadoop、Spark等知名大数据开源项目进行分析和解读,为读者开展大数据应用提供技术参考和支持。     

Underload instabilities in packet networks with flow schedulers

Instability in packet-switching networks is normally associated with overload conditions, since queueing network models show that, in simple configurations, only overload generates instability. However, some results showing that instability can happen also in underloaded queueing networks began to appear about a decade ago. Underload instabilities can be produced by: 1) customer routes that visit the same queues several times; 2) variations of the customer service times at the different queues; and 3) complex scheduling algorithms. We study, using fluid models and adversarial queueing theory, possible underload instabilities due to flow schedulers in packet networks, focusing on output queued switches with strict priority (SP) schedulers and Generalized Processor Sharing (GPS) schedulers. The considered scenarios always refer to acyclic packet routes and consider customer service times that vary only according to channel capacities, thus resembling the approaches being currently considered to provide QoS in the Internet. Our (in)stability results are rather surprising: SP schedulers appear to be more robust than GPS schedulers whenever exact information on the effective average packet flow rates is not available.     

K-PSO: An improved PSO-based container scheduling algorithm for big data applications

In recent years, Docker container technology is being applied in the field of cloud computing at an explosive speed. The scheduling of Docker container resources has gradually become a research hotspot. Existing big data computing and storage platforms apply with traditional virtual machine technology, which often results in low resource utilization, a long time for flexible scaling and expanding clusters. In this paper, we propose an improved container scheduling algorithm for big data applications named Kubernetes-based particle swarm optimization(K-PSO). Experimental results show that the proposed K-PSO algorithm converges faster than the basic PSO algorithm, and the running time of the algorithm is cut in about half. The K-PSO container scheduling algorithm and algorithm experiment for big data applications are implemented in the Kubernetes container cloud system. Our experimental results show that the node resource utilization rate of the improved scheduling strategy based on K-PSO algorithm is about 20% higher than that of the Kube-scheduler default strategy, balanced QoS priority strategy, ESS strategy, and PSO strategy, while the average I/O performance and average computing performance of Hadoop cluster are not degraded.     

The design and implementation of real-time schedulers in RED-linux

Researchers in the real-time system community have designed and studied many advanced scheduling algorithms. However, most of these algorithms have not been implemented since it is very difficult to support new scheduling algorithms on most operating systems. To address this problem, we enhance the scheduling mechanism in Linux to provide a flexible scheduling framework. In the real-time and embedded Linux (RED-Linux) project, we implement a general scheduling framework which divides the system scheduler into two components: dispatcher and allocator. The dispatcher provides the mechanism of system scheduling and resides in the kernel space. The allocator is used to define the scheduling policy and implemented as a user space function. This framework allows users to implement application-specific schedulers in the user space which is easy to program and to debug. The framework also relieves the deficiency from the stock Linux scheduler which is not designed for real-time applications. To further enhance its power, a hierarchical scheduling mechanism has been provided in RED-Linux to allow a system designer to integrate different real-time applications together. Using scheduling groups, real-time jobs can be managed and scheduled in a hierarchical manner. In this paper, we discuss how the group mechanism is implemented in RED-Linux.     

Enhancement of QoS support of HCCA schedulers using EDCA function in IEEE 802.11e networks

The IEEE 802.11e standard introduces Quality of Service support for wireless local area networks through two MAC functions: Enhanced Distributed Channel Access (EDCA) and HCF Controlled Channel Access (HCCA). While the former provides prioritized contention-based access to the medium, the latter uses a parameterized contention-free polling scheme. Several studies have proposed enhancements to EDCA or improved scheduling algorithms for HCCA to properly support VBR traffic. However, the cooperation between these functions has only marginally been considered and the solutions vary depending on specific traffic requirements.In this paper we propose a novel approach to address the problem of scheduling VBR traffic streams. Our scheduler, named Overboost, uses HCCA to negotiate a minimum bandwidth and deals with traffic streams that require more bandwidth than the negotiated one by redirecting the excess bandwidth to the EDCA function. An analytical evaluation has been conducted and the results has been corroborated by an extensive set of simulations. They show that the overall scheduler improves the performance with respect to other HCCA schedulers in terms of null rate, throughput, access delay, and queue length.     

基于TD-SCDMA的HSUPA关键技术

讨论了基于TD-SCDMA的HSUPA技术的优势及特点,以及在协议和信道结构上的改动,并详细阐述了HSUPA的各项关键技术,包括基于Node B的快速分组调度、自适应编码调制、混合自动重传请求等。最后对HSUPA技术的发展方向作了展望。     

HSUPA技术及其发展

HSUPA(高速上行分组接入)是3GPP标准化组织为进一步满足数据业务需求在WCDMA/UTRA-FDDR6版本中提出的一种新技术,本文详细描述了引入HSUPA后WCDMA系统结构的演化,阐述了HSUPA中涉及的物理层混合自动请求重传(HARQ)、快速调度、短帧长、新扩频因子以及软切换等关键技术及其相应研究,最后对HSUPA技术及WCDMA的发展趋势进行了讨论.     

On the behavior of PHM distributed schedulers for input buffered packet switches

iSLIP and parallel hierarchical matching (PHM) are distributed maximal size matching schedulers for input-buffered switches. Previous research has analyzed the hardware cost of those schedulers and their performance after a small number of iterations. In this paper, we formulate an upper bound for the number of iterations required by PHM to converge. Then, we compare the number of iterations required by iSLIP and PHM to achieve a maximal throughput under uniform Bernoulli traffic, by means of simulation. Finally, we obtain the corresponding delay performances, which are similar. The results suggest that PHM has both the advantages of previous hierarchical matching algorithms (low hardware complexity) and iSLIP (low number of iterations).     

基于Hadoop的电信大数据采集方案研究与实现

ETL是数据仓库实施过程中一个非常重要的步骤,设计一个能够对大数据进行有效处理的ETL流程以提高运营平台的采集效率,具有重要的实际意义.首先简单介绍某运营商大数据平台采集的主要数据内容.随后,为提升海量数据采集效率,提出了Hadoop与Oracle混搭架构解决方案.继而,提出一种动态触发式ETL调度流程与算法,与定时启动的ETL流程调度方式相比,可有效缩短部分流程的超长等待时间;有效避免资源抢占拥堵现象.最后,根据Hadoop和Oracle的系统运行日志,比较分析了两个平台的采集效率与数据量之间的关系.实践表明,混搭架构的大数据平台优势互补,可有效提升数据采集时效性,获得比较好的应用效果.     

Reducing computational complexity using transmitter correlation value‐assisted schedulers in multiuser MIMO uplink

In this paper, a threshold bounded antenna selection scheduler (TBS) and a computational complexity bounded antenna selection scheduler (CCBS) are proposed to reduce computational complexity in multiple input multiple output (MIMO) uplink. In contrast to previous works, a spatially correlated MIMO channel model is considered and a transmitter correlation value (TCV) is newly introduced to assist the antenna selection in addition to the channel gain. For the TBS or CCBS, with predetermined threshold of TCV or ratio of successful antennas (RSAs), full searching (FS) and sub searching (SS) are applied more efficiently to user equipments (UEs) compared with previous schedulers. As a result, the number of candidate antennas in the scheduling set can be reduced, which translates into a lower computational complexity in terms of number of evaluated antenna combinations. Additionally, compared with the TBS, the peak computational complexity can be further reduced by the CCBS. Simulation results show that with proposed schedulers the computational complexity can be reduced by at least 50% with an acceptable compromise of capacity. Copyright © 2010 John Wiley & Sons, Ltd.     

Design of packet-fair queuing schedulers using a RAM-basedsearching engine

The implementation of packet-fair queuing (PFQ) schedulers, which aim at approximating the generalized processor sharing (GPS) policy, is a central issue for providing multimedia services with various quality-of-service (QoS) requirements in packet-switching networks. In the PFQ scheduler, packets are usually time stamped with a value based on some algorithm and are transmitted with an increasing order of the time-stamp values. One of the most challenging issues is to search for the smallest time-stamp value among hundreds of thousands of sessions. In this paper, we propose a novel RAM-based searching engine (RSE) to speed up the searching process by using the concept of hierarchical searching with a tree data structure. The time for searching the smallest time stamp is independent of the number of sessions in the system and is only bounded by the memory accesses needed. The RSE can be implemented with commercial memory and field programmable gate array (FPGA) chips in a cost-effective manner. With the extension of the RSE, we propose a two-dimensional (2-D) RSE architecture to implement a general shaper-scheduler. Other challenging issues, such as time-stamp overflow and aging, are also addressed in the paper     

Best Minimum Summation Scheduler for Long Term Evolution

Resource scheduling in Long Term Evolution (LTE) is an open and rising issue. It has an enormous impact on the entire system performance. Due to the nature of LTE system, the scheduler has to be designed carefully. It has to overcome many challenges such as limited processing time and the high dynamic behavior. This paper proposes a novel scheduling policy for the MAC layer in LTE called the Best Minimum Summation (BMS). The main aim of this scheduling policy is to achieve high performance with low complexity. Three sub-schedulers have been developed. Each one of these schedulers deals with scheduling table in different dimension. The first one operates on the scheduling table through the user dimension (BMS.UE); while the second one operates on the scheduling table through the resource block dimension (BMS.RB). The third scheduler operates on the scheduling table correlating both of these dimensions (BMS.2D). All of the proposed solutions were intensively evaluated in a system level simulator. Three performance metrics were used which are throughput, error rate and fairness. The results have shown that the ability of the BMS.UE scheduler to outperform other existing schedulers of LTE.     

大数据技术与5G通信技术的融合应用

大数据技术与5G通信技术在现代社会得到了广泛的应用,两者的融合也是未来技术发展的方向之一。文中以大数据技术与5G通信技术融合应用的优势为切入点,探讨了其融合应用的方法,包括搭建融合框架、建立云端处理器、智能化信息处理等,并结合某高校信息化系统建设实例论证了上述内容。最后,文中简述了大数据技术与5G 通信技术融合应用的趋势,以服务有关技术的进一步发展。     

基于帧的ATM层次调度机制性能界及其仿真研究

对网络交换节点中所采用的调度机制的研究是网络提供QoS保证的一个重要课题.在本文中,结合基于帧调度器的简单性和GPS(generalized processor sharing)算法的良好性能,提出了一种新颖的基于帧的层次调度算法HFFQ(Hierarchical Frame-based Fair Queueing).HFFQ能同时支持实时业务和非实时业务,另外采用了简单计数的方法来替代复杂的系统虚拟时间的计算过程,大大减少了算法的实现复杂度.最后,用理论分析和仿真的方法,对HFFQ的公平性、服务率等指标的性能进行了论证.结果显示,它和PGPS(Packet-by-packet GPS)相比,在性能上有很大的提高.     

流计算大数据技术在运营商实时信令处理中的应用

基于Hadoop搭建的大数据平台采用离线批处理的方式,无法满足对数据实效性敏感的业务要求。针对运营商动态数据信息开放大数据平台的实时信令处理要求,对流式计算大数据组件进行了分析,介绍了与流计算大数据相关的实时采集、汇聚和处理组件,形成了端到端实时信令处理大数据技术解决方案,并提出了融合批处理和实时计算的大数据平台解决方案,提高了网络信令数据的时效性,为业务创新提供更大空间,带来更多利益。     

Performance impact of JobTracker failure in Hadoop

In this paper, we analyze the performance impact of JobTracker failure in Hadoop. A JobTracker failure is a serious problem that affects the overall job processing performance. We describe the cause of failure and the system behaviors because of failed job processing in the Hadoop. On the basis of the analysis, we build a job completion time model that reflects failure effects. Our model is based on a stochastic process with a node crash probability. With our model, we run simulation of performance impact with very credible failure data available from USENIX called computer failure data repository that have been collected for past 9 years. The results show that the performance impact is very severe in that the job completion time increases about four times typically, and in a worst case, it increases up to 68 times. Copyright © 2014 John Wiley & Sons, Ltd.