VRM: a failure-aware Grid resource management system

Abstract

Abstract-For resource management in Grid environments, advance reservations turned out to be very useful and hence are supported by a variety of Grid toolkits. However, failure recovery for such systems has not yet received the attention it deserves. In this paper, we address the problem of remapping reservations to other resources, when the originally selected resource fails. Instead of dealing with jobs already running, which usually means checkpointing and migration, our focus is on jobs that are scheduled on the failed resource for a specific future period of time but not started yet. The most critical factor when solving this problem is the estimation of the downtime. We avoid the drawbacks of under-or overestimating the downtime by a dynamic load-based approach that is evaluated by extensive simulations in a Grid environment and shows superior performance compared to estimationbased approaches.

Survey On Fault Tolerance In Grid Computing

“…It increases the probability of admission of the job and this planning of resources in advance allows users to gain concurrent access for their applications to be executed in parallel (i.e., in co-allocation environment). In [7] , failure recovery not only has to handle already active jobs, but also those which are admitted but not yet started, inactive jobs. This means that the affected inactive jobs have to remap in advance to another matching resource.…”

“…RMS provides uniform and transparent access to large pools of heterogeneous resources [19]. In Failure-Aware Grid resource Management system the Virtual Resource Manager (VRM) which supports QoS by means of SLAs [7]. In this work it addresses the problem of remapping reservation to other resources when the originally selected resource fails.…”

Dynamic Adaptation of Checkpoints and Rescheduling in Grid Computing

“…In Failure-Aware Grid Resource Management system [6] the Virtual Resource Manager (VRM) which supports QoS by means of SLAs. In this work it addresses the problem of remapping reservation to other resources when the originally selected resource fails.…”

Design of a Grid workflow for a climate application

“…Using elaborated reservation protocols [17] and scheduling algorithms [7,4], VRM is able to provide an SLA without requiring the local resource provider to give up their autonomy [2]. By sending only specific requests to the resources, the central ADC will never get a complete view on the local schedules.…”

“…Virtual Resource Manager (VRM) [1,3] is a Grid broker that features a mode which allows simulating the scheduling and execution of Grid workflows in a Grid. The Grid model can be derived from a real world installation or a synthetic setup.…”