We propose the implementation of a distributed system for DL storage and retrieval which relies on two earlier developed technologies: buckets and document linking base on user retrieval patterns. Buckets are expanded...
详细信息
We propose the implementation of a distributed system for DL storage and retrieval which relies on two earlier developed technologies: buckets and document linking base on user retrieval patterns. Buckets are expanded with a mechanism to keep track of weighted bucket relations. Each bucket keeps track of its connections to other related buckets. A mechanism is put in place which allows buckets to change their connections to other buckets according to the registration of bucket retrieval patterns issued by users. The collection of buckets thereby dynamically updates its link structure according to user retrieval patterns.
Clusters of workstations (COW) offer high performance relative to their cost. Generally these clusters operate as autonomous systems running independent copies of the operating system, where access to machines is not ...
详细信息
Clusters of workstations (COW) offer high performance relative to their cost. Generally these clusters operate as autonomous systems running independent copies of the operating system, where access to machines is not controlled and all users enjoy the same access privileges. While these features are desirable and reduce operating costs, they create adverse effects on parallel applications running on these clusters. Load imbalances are common for parallel applications on COWs due to: 1) variable amount of load on nodes caused by an inherent lack of parallelism, 2) variable resource availability on nodes, and 3) independent scheduling decisions made by the independent schedulers on each node. Our earlier study has shown that an approach combining static program analysis, dynamic load balancing, and scheduler cooperation is effective in countering the adverse effects mentioned above. In our current study, we investigate the scalability of our approach as the number of processors is increased. We further relax the requirement of global synchronization, avoiding the need to use barriers and allowing the use of any other synchronization primitives while still achieving dynamic load balancing. The use of alternative synchronization primitives avoids the inherent vulnerability of barriers to load imbalance. It also allows load balancing to take place at any point in the course of execution, rather than only at a synchronization point, potentially reducing the time the application runs imbalanced. Moreover, load readjustment decisions are made in a distributed fashion, thus preventing any need for processes to globally synchronize in order to redistribute load.
resourcemanagement in the typical Grid environment based on multi-MPP systems or clusters today still is one of the challenging problems. We will present MeSch, a solution for the problem of resource allocation and j...
详细信息
ISBN:
(纸本)3540414037
resourcemanagement in the typical Grid environment based on multi-MPP systems or clusters today still is one of the challenging problems. We will present MeSch, a solution for the problem of resource allocation and job scheduling in a distributed heterogeneous environment. MeSch has been implemented and tested successfully in the heterogeneous multi-MPP environment of GMD's Institute for Scientific Computing and Algorithms. MeSch allows users to access simultaneously, through a single request, heterogeneous resources distributed across the linked systems. This is possible either through explicit demands for different resources or through implicit scheduling of resources resulting from interpretation of requests. The scheduling system is available for both batch and interactive usage of resources. MeSch is implemented based on locally available scheduling facilities thus respecting the different schedulingsystems and policies of the computing centers in the Grid.
Scientific and other grand challenge applications are a driving force for developing computing infrastructure of the future. Their constantly increasing computational power requirements often cannot be met by availabl...
详细信息
ISBN:
(纸本)0769510116
Scientific and other grand challenge applications are a driving force for developing computing infrastructure of the future. Their constantly increasing computational power requirements often cannot be met by available systems [1]. Such a situation has led to the emergence of new platform called "Computational Grid" [2]. It seems very important to identify, common and reusable components that are needed and can be used in a computational grid by different applications. Stich components could be the Grid information services, security mechanisms, resourcemanagement, large data set management etc. This paper sketches the issues of one of the most important components of the grid - the resourcemanagement and scheduling component. We will present our work on multiobjective resourcemanagement system. The novelty of the approach lays in its multiobjective nature. We present the Multi-Criteria resource Broker (MC-Broker), whose scheduling mechanisms are driven by the user preferences regarding various scheduling criteria, such as computation time and cost, communication time between distributed processes, level of load balancing and others.
Computational Grids are a promising platform for executing large-scale resource intensive applications. However, resourcemanagement and scheduling in the Grid environment is a complex undertaking as resources are (ge...
详细信息
We address the problem of Quality of Service (QoS) requirements for multimedia applications (e.g., distance education, telemedicine, electronic commerce). These applications need to be able to co-exist with more tradi...
详细信息
ISBN:
(纸本)3540416102
We address the problem of Quality of Service (QoS) requirements for multimedia applications (e.g., distance education, telemedicine, electronic commerce). These applications need to be able to co-exist with more traditional applications for transaction and data processing and have soft real-time requirements. Unlike most other work in QoS management, we provide a framework that does not require users or application developers to have detailed knowledge of the resources needed and resourcescheduling and allocation techniques in use. These underlying details are effectively hidden. In this paper, we describe our strategy, an architecture of services to support the strategy and a prototype.
The proceedings contain 17 papers. The special focus in this conference is on Object Middleware, resource Discovery and management. The topics include: Grid application design using software components and web service...
ISBN:
(纸本)3540429492
The proceedings contain 17 papers. The special focus in this conference is on Object Middleware, resource Discovery and management. The topics include: Grid application design using software components and web services;design and implementation of a CORBA commodity grid kit;towards high performance CORBA and MPI middlewares for grid computing;an integrated grid environment for component applications;high performance knowledge discovery on the grid;on fully decentralized resource discovery in grid environments;an adaptive service grid architecture using dynamic replica management;identifying dynamic replication strategies for a high-performance datagrid;resource co-allocation on the computational grid;a computational grid-wide queuing system;a scheduling model for grid computing systems;encapsulated approaches to grid service architecture;a methodology for account management in grid computing environments;a framework for authorization, accounting policy specification and evaluation in grids;predicting and monitoring grid application behavior;production-level distributed parametric study capabilities for the grid and the DO experiment data grid.
When users' tasks in a distributed heterogeneous computing environment are allocated resources, and the total demand placed on system resources by the tasks, for a given interval of time, exceeds the resources ava...
详细信息
暂无评论