A prediction-based dynamic file assignment strategy for parallel file systems

Nowadays, the rapid development of the internet calls for a high performance file system, and a lot of efforts have already been devoted to the issue of assigning nonpartitioned files in a parallel file system with the aim of pursuing a prompt response to requests. Yet most of the existing strategies still fail to bring about an optimal performance on system mean response time metrics, and new strategies which can achieve better performance in terms of mean response time become indispensable for parallel file systems. This paper, while addressing the issue of assigning nonpartitioned files in parallel file systems where the file accesses exhibit Poisson arrival rates and fixed service times, presents an on-line file assignment strategy, named prediction-based dynamic file assignment (PDFA), to minimize the mean response time among disks under different workload conditions, and a comparison of the PDFA with the well-known file assignment algorithms, such as HP and SOR. Comprehensive experimental results show that PDFA is able to improve the performance consistently in terms of mean response time among all algorithms for comparison.     

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。     

Fault tolerant file models for parallel file systems: introducing distribution patterns for every file

Parallelism in file systems is obtained by using several independent server nodes supporting one or more secondary storage devices. This approach increases the performance and scalability of the system, but a fault in one single node can stop the whole system. To avoid this problem, data must be stored using some kind of redundant technique, so any data stored in a faulty element can be recovered. Fault tolerance can be provided in I/O systems by using replication or RAID based schemes. However, most of the current systems apply the same technique for all files in the system. This paper describes the fault tolerance support provided by Expand, a parallel file system based on standard servers. This support can be applied to other parallel file systems with many benefices: fault tolerance at file level, flexible definition of fault tolerance scheme to be used, possibility to change the fault tolerant support used for a file, etc.

Multidimensional array I/O in Panda 1.0

Large multidimensional arrays are a common data type in high-performance scientific applications. Without special techniques for handling input and output, I/O can easily become a large fraction of execution time for applications using these arrays, especially on parallel platforms. Our research seeks to provide scientific programmers with simpler and more abstract interfaces for accessing persistent multidimensional arrays, and to produce advanced I/O libraries supporting more efficient layout alternatives for these arrays on disk and in main memory. We have created the Panda (Persistence AND Arrays) I/O library as a result of developing interfaces and libraries for applications in computational fluid dynamics in the areas of checkpoint, restart, and time-step output data. In the applications we have studied, we find that a simple, abstract interface can be used to insulate programmers from physical storage implementation details, while providing improved I/O performance at the same time.(A preliminary version of this paper was presented at Supercomputing '94.)     

A dynamic and adaptive load balancing strategy for parallel file system with large-scale I/O servers

Many solutions have been proposed to tackle the load imbalance issue of parallel file systems. However, all these solutions either adopt centralized algorithms, or lack considerations for both the network transmission and the tradeoff between benefits and side-effects of each dynamic file migration. Therefore, existing solutions will be prohibitively inefficient in large-scale parallel file systems. To address this problem, this paper presents SALB, a dynamic and adaptive load balancing algorithm which is totally based on a distributed architecture. To be also aware of the network transmission, SALB on the one hand adopts an adaptively adjusted load collection threshold in order to reduce the message exchanges for load collection, and on the other hand it employs an on-line load prediction model with a view to reducing the decision delay caused by the network transmission latency. Moreover, SALB employs an optimization model for selecting the migration candidates so as to balance the benefits and the side-effects of each dynamic file migration. Extensive experiments are conducted to prove the effectiveness of SALB. The results show that SALB achieves an optimal performance not only on the mean response time but also on the resource utilization among the schemes for comparison. The simulation results also indicate that SALB is able to deliver high scalability.     

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.     

Filter Cache:一种提高Lustre I/O性能的方法

高性能计算系统需要一个可靠高效的并行文件系统.Lustre集群文件系统是典型的基于对象存储的集群文件系统,它适合大数据量聚合I/O操作.大文件I/O操作能够达到很高的带宽,但是小文件I/O性能低下.针对导致Lustre的设计中不利于小文件I/O操作的两个方面,提出了Filter Cache方法.在Lustre的OST组件中设计一个存放小文件I/O数据的Cache,让OST端的小文件I/O操作异步进行,以此来减少用户感知的小文件I/O操作完成的时间,提高小文件I/O操作的性能.     

并行I/O中一种基于服务时间最小变化的文件分配策略

论文中对于文件访问的服务时间进行了较深入的研究,提出一种并行I/O文件分配算法——启发式文件分类分配策略,它在负载基本均衡前提下,按照相似的访问服务时间对每个待分配的数据文件进行磁盘分配。通过对启发式文件分类分配策略与已有的贪婪文件分配法进行实验比较,结果表明:系统处理重负载时,访问响应时间提高了30%左右,而且数据访问速率越高,由启发式文件分类分配策略所提高的性能就越明显。     

I/O performance of an RAID-10 style parallel file system

Without any additional cost, all the disks on the nodes of a cluster can be connected together through CEFT-PVFS, an RAID-10 style parallel file system, to provide a multi-GB/s parallel I/O performance.I/O response time is one of the most important measures of quality of service for a client. When multiple clients submit data-intensive jobs at the same time, the response time experienced by the user is an indicator of the power of the cluster. In this paper, a queuing model is used to analyze in detail the average response time when multiple clients access CEFT-PVFS. The results reveal that response time is with a function of several operational parameters. The results show that I/O response time decreases with the increases in I/O buffer hit rate for read requests, write buffer size for write requests and the number of server nodes in the parallel file system, while the higher the I/O requests arrival rate, the longer the I/O response time. On the other hand, the collective power of a large cluster supported by CEFT-PVFS is shown to be able to sustain a steady and stable I/O response time for a relatively large range of the request arrival rate.     

Network and device-level impacts: performance and reliability of active I/O storage systems

Distributed active storage architectures are designed to offload user-level processing to the peripheral from the host servers. In this paper, we report preliminary investigation on performance and fault recovery designs, as impacted by emerging storage interconnect protocols and state-of-the-art storage devices. Empirical results obtained using validated device-level and interconnect data demonstrate the significance of the said parameters on the overall system performance and reliability.     

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.     

并行文件系统扩容与I/O节点负载平衡策略

在规模并行处理系统ＭＰＰ商业化应用中,并行文件系统容量与Ｉ／Ｏ节点的负载平衡一直是挖掘整个系统并行性,进而提高系统性能的关键问题之一。本文从分析Ｐａｒａｇｏｎ大规模并行处理系统的体系结构和并行文件系统的体系结构出发,给出了并行文件系统的扩容策略,使用ＭＰＰ系统的Ｉ／Ｏ节点平衡负载理论模型,给出了实际工程应用中的负载平衡策略。     

集群计算并行I/O中动态文件分配策略

针对集群计算系统中并行I／O文件分配问题进行了研究,提出一种新的动态文件分配算法一启发式负载与服务时间变化综合平衡策略。该策略将所需要的访问服务时间相似的文件分配到集群系统的同一结点磁盘上,并保证每个磁盘上的负载值不超过负载平衡的极限值,从而在所有结点磁盘间既基本实现负载均衡,又使每个磁盘上文件间的服务时间变化最小化。实验结果表明了该策略在性能提高方面的有效性。     

Exploring optimal combination of a file system and an I/O scheduler for underlying solid state disks

Performance and energy consumption of a solid state disk （SSD） highly depend on file systems and I/O schedulers in operating systems. To find an optimal combination of a file system and an I/O scheduler for SSDs, we use a metric called the aggregative indicator （AI）, which is the ratio of SSD performance value （e.g., data transfer rate in MB/s or throughput in IOPS） to that of energy consumption for an SSD. This metric aims to evaluate SSD performance per energy consumption and to study the SSD which delivers high performance at low energy consumption in a combination of a file system and an I/O scheduler. We also propose a metric called Cemp to study the changes of energy consumption and mean performance for an Intel SSD （SSD-I） when it provides the largest AI, lowest power, and highest performance, respectively. Using Cemp, we attempt to find the combination of a file system and an I/O scheduler to make SSD-I deliver a smooth change in energy consumption. We employ Filebench as a workload generator to simulate a wide range of workloads （i.e., varmail, fileserver, and webserver）, and explore optimM combinations of file systems and I/O schedulers （i.e., optimal values of AI） for tested SSDs under different workloads. Experimental results reveal that the proposed aggregative indicator is comprehensive for exploring the optimal combination of a file system and an I/O scheduler for SSDs, compared with an individual metric.     

并行文件系统Lustre细粒度I/O性能优化

并行文件系统Lustre粗粒度I/O性能良好,细粒度I/O性能相对粗粒度I/O比较低下,因此优化细粒度I/O性能成为提高系统整体I/O性能的关键问题。在研究和分析了Lustre的I/O访问模式、细粒度I/O服务流程和页面替换算法等方面后,提出了细粒度优先（Fine Grained First,FGF）LRU算法。在OST端及Client端的页高速缓存中最大程度地保留细粒度I/O的页面,降低细粒度I/O引起的页面下沉速度,延长细粒度I/O页面在主存中的时间,进而减少对磁盘的访问次数,降低磁盘访问开销。通过对实验数据的对比和分析,验证了FGF-LRU算法的有效性。在不影响粗粒度I/O性能的情况下,提高了细粒度I/O性能,最终实现提高系统整体I/O性能。     

一种高性能I／O方法

并存文伴系统是解决I／O瓶颈问题的重要途径。研究表明，科学应用中跨越式的文件访问模式与现存并行文件系统访问这些数据的方法的结合，对于大型数据集的访问其I／O性能是难以接受的。为了提高并行文件系统中对不连续数据的I／O性能，创建了一种新型高性能I／O方法：用户自定义文件视图结合合并I／O请求。并且在WPFS并行文件系统中实现了该方法。研究和实验结果表明，该方法具有增强科学应用性能的潜力。     

并行数据库系统中并行I／O的研究与实现

并行Ｉ／Ｏ是基于无共享结构的并行数据库系统提高性能的有效途径之一。它通过并行磁盘服务和网络传输并行化提供了高带宽Ｉ／Ｏ。文章设计实现了基于无共享结构的并行数据库系统的并行Ｉ／Ｏ,探讨了设计并行Ｉ／Ｏ时的几个关键问题及实现技术。     

并行文件系统与并行I／O研究

集群计算系统中并行文件系统的研究是当前计算机与网络技术中的一个热点问题,而并行I/O是缓解系统数据输入输出瓶颈的一个技术途径。论文对当前集群系统中的并行文件系统与并行I/O做了研究,阐述了研究发展的现状、关键问题等,指出了在集群计算系统中的文件组织、分布以及其在磁盘上的实现、数据的访问特性、高性能网络文件系统、系统的负载平衡与缓冲和预取策略。     

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。     

并行存储反应调度算法研究

文章提出了一种并行存储反应调度算法,它是基于存储访问建模的和基于规则的存储自动优化算法。这种算法使用E_IS_PPM和Last_N_Successor算法对存储访问建模,然后对存储访问模式进行分类,并确立了存储优化的规则集。最后,在MPI基础上实现了调度算法。