highly available metadata services of distributed file systems are essential to cloud applications. However, existing highly available metadata designs lack client-oriented features that treat metadata discriminately,...
详细信息
ISBN:
(纸本)9781467380119
highly available metadata services of distributed file systems are essential to cloud applications. However, existing highly available metadata designs lack client-oriented features that treat metadata discriminately, leading to a single metadata fault domain and low availability. After investigating the workload characteristics of Hadoop, we propose Client-Oriented METadata (COMET), a novel highly available metadata service design that divides and distributes metadata into independent regions in terms of clients. these regions are isolated fault domains inherently, and failures in one region will not break file operations in other regions. A prototype of COMET was implemented based on HDFS, and the experimental results show that COMET can significantly improve metadata availability of HDFS without obvious performance degradation. It can also deliver scalable performance and faster metadata recovery due to its decentralized architecture.
Fault-tolerant scheduling is an imperative step for large-scale computational Grid systems, as often geographically distributed nodes co-operate to execute a task. By and large, the primary-backup approach is a common...
详细信息
ISBN:
(纸本)1595936734
Fault-tolerant scheduling is an imperative step for large-scale computational Grid systems, as often geographically distributed nodes co-operate to execute a task. By and large, the primary-backup approach is a common methodology used for fault tolerance where in each task has a primary copy and a backup copy on two different processors. Backup overloading has been proposed to reduce replication cost by allowing the backup copy to overload with other backup copies on the same processor. In this paper, we consider two classes of independent tasks where in boththe classes have fault-tolerance requirements. Furthermore, Class 1 tasks require the response time to be as short as possible when a fault occurs, while Class 2 tasks prefer backups with minimum replication cost. We propose two algorithms, called the MRC-ECT algorithm and the MCT-LRC algorithm. Algorithm MRC-ECT is shown to guarantee an optimal backup schedule in terms of replication cost, while MCT-LRCcan schedule a backup with minimum completion time and low replication cost. We conduct extensive simulation experiments to quantify the performance of the proposed algorithms.
Given recent algorithm, software, and hardware innovation, computing has enabled a plethora of new applications. As computing becomes increasingly ubiquitous, however, so does its environmental impact. this paper brin...
详细信息
ISBN:
(纸本)9781665422352
Given recent algorithm, software, and hardware innovation, computing has enabled a plethora of new applications. As computing becomes increasingly ubiquitous, however, so does its environmental impact. this paper brings the issue to the attention of computer-systems researchers. Our analysis, built on industry-reported characterization, quantifies the environmental effects of computing in terms of carbon emissions. Broadly, carbon emissions have two sources: operational energy consumption, and hardware manufacturing and infrastructure. Although carbon emissions from the former are decreasing thanks to algorithmic, software, and hardware innovations that boost performance and power efficiency, the overall carbon footprint of computer systems continues to grow. this work quantifies the carbon output of computer systems to show that most emissions related to modern mobile and data-center equipment come from hardware manufacturing and infrastructure. We therefore outline future directions for minimizing the environmental impact of computing systems.
In Internet-scale distributed and replicated services, poor consistency results in poor QoS or even monetary loss. Recent research focuses on enforcing a certain consistency level, instead of perfect consistency, to s...
详细信息
ISBN:
(纸本)1595936734
In Internet-scale distributed and replicated services, poor consistency results in poor QoS or even monetary loss. Recent research focuses on enforcing a certain consistency level, instead of perfect consistency, to strike a balance between consistency guarantee and system's scalability. In this paper, we argue that it is equally, if not more, important to achieve adaptability. I.e., the system adjusts its consistency level on the fly to suit applications. ongoing need. this paper presents IDEA (an Infrastructure for DEtection-based Adaptive consistency control), a protocol that adaptively controls consistency in replicated services by detecting inconsistency among nodes in a timely manner via an inconsistency detection framework and resolving the detected inconsistencies efficiently when necessary. through experimentation on Planet-Lab, IDEA is evaluated from two aspects: its adaptive interface and its performance of inconsistency resolution. Results show that IDEA achieves adaptability by adjusting the consistency level according to users. preference on-demand, and it achieves low inconsistency resolution delay and incurs minimal communication cost.
Linear Hashing is a widely used and efficient version of extensible hashing. A distributed version of Linear Hashing is LH* that stores key-indexed records on up to hundreds of thousands of sites in a distributed syst...
详细信息
ISBN:
(纸本)9781479929276
Linear Hashing is a widely used and efficient version of extensible hashing. A distributed version of Linear Hashing is LH* that stores key-indexed records on up to hundreds of thousands of sites in a distributed system. LH* implements the dictionary data structure efficiently since it does not use a central component for the key-based operations of insertion, deletion, actualization, and retrieval and for the scan operation. LH* allows a client or a server to commit an addressing error by sending a request to the wrong server. In this case, the server forwards to the correct server directly or in one more forward operation. We discuss here methods to avoid the double forward, which is rare but might breach quality of service guarantees. We compare our methods with LH* P2P that pushes information about changes in the file structure to clients, whether they are active or not.
Traditional financial trading systems were developed in database-based standalone model years ago. Given that there exist some evident drawbacks like high latency and low availability in such architecture, it is imper...
详细信息
ISBN:
(纸本)9781479961238
Traditional financial trading systems were developed in database-based standalone model years ago. Given that there exist some evident drawbacks like high latency and low availability in such architecture, it is imperative to design a new model which can meet the requirements of time-sensitive financial transactions. therefore, in this paper we design and implement a state-replication-based matching system which includes the critical technical points like order sequencing and asynchronous data persistence. the performance experiment indicates that this new model can dramatically increase the throughput by three times and the latency only accounts for nearly 10% compared withthe conventional architecture. high availability and robustness can be guaranteed as well.
We propose in this paper a new algorithm that, when called by existing token ring-based algorithms of parallel and distributed applications, easily renders the token tolerant to losses in presence of node crashes. At ...
详细信息
ISBN:
(纸本)9781479929276
We propose in this paper a new algorithm that, when called by existing token ring-based algorithms of parallel and distributed applications, easily renders the token tolerant to losses in presence of node crashes. At most k consecutive node crashes are tolerated in the ring. Our algorithm scales very well since a node monitors the liveness of at most k other nodes and neither a global election algorithm nor broadcast primitives are used to regenerate a new token. It is thus very effective in terms of latency cost. Finally, a study of the probability of having at most k consecutive node crashes in the presence of f failures and a discussion of how to extend our algorithm to other logical topologies are also presented.
Resource sharing and implementation of software stack for emerging multicore processors introduce performance and scaling challenges for large-scale scientific applications, particularly on systems withthousands of p...
详细信息
ISBN:
(纸本)9780769534237
Resource sharing and implementation of software stack for emerging multicore processors introduce performance and scaling challenges for large-scale scientific applications, particularly on systems withthousands of processing elements. Traditional performance optimization, tuning and modeling techniques that rely on uniform representation of computation and communication requirements are only partially useful due to the complexity of applications and underlying systems and software architecture. In this paper, we propose a workload modeling methodology that allows application developers to capture and represent hierarchical decomposition and distribution of their applications thereby allowing them to explore and identify optimal mapping of a workload on a target system. We demonstrate the proposed methodology on a Teraflopsscale fusion application that is developed using message-passing (MPI) programming paradigm. Using our analysis and projection results, we obtain insight into the performance characteristics of the application on a quad-core system and also identify optimal mapping on a Teraflops-scale platform.
this paper describes architectural and performance aspects of the digital computer control system used for the PBX-M Plasma Control System (PPCS). the goal of the PPCS is to achieved integrated and improved plasma con...
详细信息
ISBN:
(纸本)0780329708
this paper describes architectural and performance aspects of the digital computer control system used for the PBX-M Plasma Control System (PPCS). the goal of the PPCS is to achieved integrated and improved plasma control. Integration consists of replacing control functions presently served by several analog systems with a realtime digital control system. the inherently dynamic control capabilities of a highperformance digital system foster exploration of advanced plasma control concepts to serve future tokamaks. the PPCS will run concurrent multiple feedback control loops, with input, processing, and output times ranging from 100 μS to 10 milliseconds. the initial control loop for plasma shaping was expected to complete in approximately 300 μS. the VME-based realtime computing hardware will be described. In addition, measurements of the system's performance such as data i/o rates and computingperformance will be shown. the information presented herein covers the results of the computer system's design, configuration, and laboratory testing. Actual plasma control has not been accomplished to date.
While industry is making rapid advances in system virtualization, for server consolidation and for improving system maintenance and management, it has not yet become clear how virtualization can contribute to the perf...
详细信息
ISBN:
(纸本)1595936734
While industry is making rapid advances in system virtualization, for server consolidation and for improving system maintenance and management, it has not yet become clear how virtualization can contribute to the performance of high end systems. In this context, this paper addresses a key issue in system virtualization - how to efficiently virtualize I/O subsystems and peripheral devices. We have developed a novel approach to I/O virtualization, termed self-virtualized devices, which improves I/O performance by off loading select virtualization functionality onto the device. this permits guest virtual machines to more efficiently (i.e., with less overhead and reduced latency) interact withthe virtualized device. the concrete instance of such a device developed and evaluated in this paper is a self-virtualized network interface (SV-NIC), targeting the high end NICs used in thehighperformance domain. the SV-NIC (1) provides virtual interfaces (VIFs) to guest virtual machines for an underlying physical device, the network interface, (2) manages the wayin which the device's physical resources are used by guest operating systems, and (3) provides highperformance, low overhead network access to guest domains. Experimental results are attained in a prototyping environment using an IXP 2400-based ethernet board as a programmable network device. the SV-NIC scales to large numbers of VIFs and guests, and offers VIFs with 77% higher throughput and 53% less latency compared to the current standard virtualized device implementations on hyper visor-based platforms. Copyright 2007 ACM.
暂无评论