distributed checkpointing is an important concept in providing fault tolerance in distributed systems. In today's applications, e.g., grid and massively parallel applications, the imposed overhead of taking a dist...
详细信息
ISBN:
(纸本)0769523315
distributed checkpointing is an important concept in providing fault tolerance in distributed systems. In today's applications, e.g., grid and massively parallel applications, the imposed overhead of taking a distributed checkpoint using the known approaches can often outweigh its benefits due to coordination and other overhead from the processes. This paper presents an innovative approach for distributed checkpointing. In this approach, the checkpoints are obtained using offline analysis based on the application level. During execution, no coordination is required. After presenting our approach, we prove its safety and present a performance analysis of it using stochastic models.
Numerous studies have been targeting the problem of scheduling divisible workloads in distributedcomputing environments. The UMR (Uniform Multi-Round) algorithm stands out from all others by being the first close-for...
详细信息
ISBN:
(纸本)9780889866386
Numerous studies have been targeting the problem of scheduling divisible workloads in distributedcomputing environments. The UMR (Uniform Multi-Round) algorithm stands out from all others by being the first close-form optimal scheduling algorithm. However, present algorithms, including the UMR, do not pay due attention to optimizing the set of workers that get selected to participate in processing workload chunks. In addition to the absence of a good resource selection policy, the UMR relies primarily in its computation on the CPU speed and overlooks the role of other key parameters such as network bandwidth. In this paper, we propose an extended version of UMR, called UMR2, that overcomes these limitations and adopts a worker selection policy that aims at minimizing the makespan. We, theoretically and experimentally, show that UMR2 is superior to UMR, specifically in a WAN computing platform such as the grid.
Large-scale online applications such as Massively Multiplayer Online Games (MMOGs) require large amount of computing resources that support many players interacting simultaneously. Cluster computing is the technology ...
详细信息
ISBN:
(纸本)9781424412501
Large-scale online applications such as Massively Multiplayer Online Games (MMOGs) require large amount of computing resources that support many players interacting simultaneously. Cluster computing is the technology mostly used by online game designing firms. Cluster computing is limited by the number and types of computers it can manage, but these computers are usually in the same geographical location. On the other hand, gridcomputing offers large-scale high performance distributedcomputing which connects various types of computing resources on the Internet. In this paper, we design a gridcomputing platform called the Massively Multi-user Online Platform (MMOP). The objectives of this proposed design are to offer scalability, flexibility, and simplicity to the development processes of distributed applications. MMOP allows executions of applications based on specified policy rules with dynamic addition of computing resources at run-time. Each application is managed separately, and multiple. large-scale applications can share a single computing architecture. An online game has been built to test the functional behavior of the MMOP. From the simulation results, the MMOP has demonstrated as a high performance and scalable computing architecture.
gridcomputing using idle desktop PCs is popular as a low-cost platform for high-throughput applications. However, they require high-cost data transfer systems from file servers to calculating nodes when very large am...
详细信息
ISBN:
(纸本)9780889866386
gridcomputing using idle desktop PCs is popular as a low-cost platform for high-throughput applications. However, they require high-cost data transfer systems from file servers to calculating nodes when very large amounts of data are handled. This high cost derives from inexpensive I/O of worker PCs and small inexpensive file systems of desktop gridcomputing. This work describes a desktop gridcomputing system and proposes a resource management system for volume visualization. The resource management system proposed in this work overcomes problems with high-cost data transfer and helps us to use desktop gridcomputing for data-intensive applications. The resource management system provides efficient scheduling while considering the data file location and consists of three technical parts: resource management, multiple data replication, and worker selection. The proposed system uses these techniques to make an intelligent reuse of data which has been previously copied. This paper evaluates the proposed system compared with a traditional system that does not consider data file location. The proposed resource management System provides efficient scheduling and appears to achieve good performance for data-intensive applications.
The current efforts on programming grid applications often rely on service-oriented approaches like grid Services. This paper presents HOC-SA - a Service Architecture for Higher-Order Components, which provides the pr...
详细信息
ISBN:
(纸本)0769522254
The current efforts on programming grid applications often rely on service-oriented approaches like grid Services. This paper presents HOC-SA - a Service Architecture for Higher-Order Components, which provides the programmer with reusable and composable patterns of parallelism and is interoperable with the latest Globus Toolkit implementations. We describe our implementation of HOC-SA using OGSA-DAI, a framework for integrating grids with distributed databases. We present a simple example application and report first measurements on our grid testbed.
grid is a promising computing platform that integrates resources from different organizations in a shared, coordinated and collaborative manner to solve large-scale science and engineering issues. grid technologies ar...
详细信息
ISBN:
(纸本)1892512459
grid is a promising computing platform that integrates resources from different organizations in a shared, coordinated and collaborative manner to solve large-scale science and engineering issues. grid technologies are evolving towards an Open grid Services Architecture (OGSA). The OGSA provides a uniform service-oriented architecture and integrates grid technologies with emerging Web services standards. We are constructing an OGSA-based gridcomputing platform - Science and Engineering Research grid (SER-grid). This platform is built on Globus 2.2, the de facto gridcomputing standard, and integrated with Web service technologies. All the grid services in this platform are viewed as Web services to forma gridcomputing environment. This paper describes the basic requirements and design of this gridcomputing platform, including the essential components of the platform and the key grid services provided We present the architecture of two key grid services, i.e. Information Service and Execution Management Service. The implementation issues of this gridcomputing platform are also discussed.
The worlds' energy demand keeps increasing. To address this demand of energy growth intelligent energy management, monitoring and controlling play a big role besides increasing the energy generation capacity. The ...
详细信息
ISBN:
(纸本)9781479965571
The worlds' energy demand keeps increasing. To address this demand of energy growth intelligent energy management, monitoring and controlling play a big role besides increasing the energy generation capacity. The future smart grid expected to accommodate and integrates widely dispersed energy sources and distributed energy generations. Advanced information and communication technologies, smart meters and sensors are implemented in smart grid to enable an automated, intelligent and widely distributed energy delivery network. A huge amount of row data is collected by smart meters and sensors from the end user and different part of the network to the computation system, monitoring and controlling centers. An enormous pool of computing resources and storage must be provided to compute this vast amount of data. Processing these data enables the operators to make a knowledge based action as it occurs. This paper discusses the monitoring and controlling of distributed renewable energy generations in smart grid using cloud computing resources. Design and develop a Lab demonstrator and realization of the proposed solution to utilize the advantage of unlimited resources provided by cloud computing.
Todays scientific applications have huge data requirements which continue to increase drastically every year These data are generally accessed by many users from all across the the globe. This implies a major necessit...
详细信息
ISBN:
(纸本)0769520863
Todays scientific applications have huge data requirements which continue to increase drastically every year These data are generally accessed by many users from all across the the globe. This implies a major necessity to move huge amounts of data around wide area networks to complete the computation cycle, which brings with it the problem of efficient and reliable data placement. The current approach to solve this problem of data placement is either doing it manually, or employing simple scripts which do not have any automation or fault tolerance capabilities. Our goal is to make data placement activities first class citizens in the grid just like the computational jobs. They will be queued, scheduled, monitored, managed, and even check-pointed. More importantly, it will be made sure that they complete successfully and without any human interaction. We also believe that data placement jobs should be treated difterently from computational jobs, since they may have different semantics and different characteristics. For this purpose, we have developed Stork, a scheduler for data placement activities in the grid.
The distributed query is one of the research focus in the Big Data. Nowadays, many companies and institutions provide technology and products to realize function or improve efficiency in the all kinds of database. In ...
详细信息
ISBN:
(纸本)9781479966219
The distributed query is one of the research focus in the Big Data. Nowadays, many companies and institutions provide technology and products to realize function or improve efficiency in the all kinds of database. In the scene of electricity, using these techniques, the real-time requirement (<10s) cannot be met. This paper provides a real-time distributed query solution with Objectification Parallel computing (OPC) to solve the above challenges. The data split from Big Data, is distributed stored in memory of cluster in the OPC. In the solution, making use of the thought of divide and rule and tree merging, there are two stages. The first stage is local data query. The intermediate query result can be obtained. The second stage is multistage summarizing. The final result can be returned to user. The solution has been applied to the power production management system (PMS) of State grid of China. The results show that solution is efficiently reliable and meets real-time.
Data-intensive grid applications need access to large datasets that may each be replicated on different resources. Minimizing the overhead of transferring these datasets to the resources where the applications are exe...
详细信息
ISBN:
(纸本)9781424403431
Data-intensive grid applications need access to large datasets that may each be replicated on different resources. Minimizing the overhead of transferring these datasets to the resources where the applications are executed requires that appropriate computational and data resources be selected. In this paper, we introduce a heuristic for the selection of resources based on a solution to the Set Covering Problem (SCP). We then pair this mapping heuristic with the well-known MinMin scheduling algorithm and conduct performance evaluation through extensive simulations.
暂无评论