A Network Bandwidth-Aware Job Scheduling with Dynamic Information Model for Grid Resource Brokers

Abstract

In this paper, we propose a resource broker, which providing a friendly interface for accessing available and appropriate resources via user credentials, is developed on a platform constructed by employing the Globus toolkit. This broker not only deploys a domainbased network information model and its dynamic version to measure network status by invoking Network Weather Service (NWS) on grid computing environments, but also uses the Ganglia and NWS tools to monitor resource status and network-related information. Based on the actual value of network bandwidth and preserving the advantages of our previous model, a network bandwidth-award job scheduling algorithm for grid resource brokerage is then developed for communication-intensive job execution. Furthermore, the network information and resource status collected are all up-to-date so that the resource broker can effectively match appropriate grid resources and users' requests to improve execution.

Resource brokering using a multi‐site resource allocation strategy for computational grids

“…Two services, the Information Service and the Monitoring Service, were retained to achieve this. Figure 4 depicts the Information Service, Monitoring Service, and Resource Broker architectures of our previous version [44,47]. Unlike It employs the Ganglia and the NWS to collect resource information and current network bandwidth.…”

“…The only way to obtain the information between them is by estimation. To resolve this drawback, we propose a model, called the dynamic domain-based network information model [47], which as shown in Figure 3 dynamically rotates the role of the site head node to the next node when the head node is busy, where busy is defined as that where the load of the node is over a threshold, e.g. 80%.…”

“…Our previous link-bandwidth measurement scheme [44,47] measures a total of NMN(n) = n ×(n −1) network links per T seconds, producing a huge amount of measurement traffic and degrading system performance where n is the number of nodes in the Grid of concern. One of our previous works [11] proposed the domain-based network information model which is a static link-measure model.…”