Youngjoo Han

A new grid resource management mechanism with resource-aware policy administrator for SLA-constrained applications

Due to the dynamic nature and an uncertainty of Grid computing, system behavior in terms of Quality of Service (QoS) depends crucially on the selection of appropriate subset of the available resources. Thus, a well-organized system architecture that provides high system availability using a scheduling mechanism for Grid applications is required. In this paper, we design an efficient architecture referred to as Resource-Aware Policy Administrator (RAPA) with Service Level Agreement (SLA) and real-time attributes in Grid resources and propose a new Grid resource management mechanism with an admission control scheme to make an optimal resource allocation policy for both satisfying QoS requirements in terms of deadline as well as profit and accepting more jobs. Also, we demonstrate detailed workings of the proposed schemes with the RAPA via a simulation study by using real-life parameters. The simulation results show that the proposed schemes have an effect on accepting more jobs with satisfying the deadline of jobs and increasing service profits and can guarantee high system availability in terms of system performance.

SLA-Constrained Policy-Based Scheduling Mechanism in Grid

Due to the dynamic nature and uncertainty of grid computing, system rehability can become very unpreaiciaDie. inus, a well-defined scheduling mechanism that provides high system availability for grid applications is required. In this letter, we propose a SLA-constrained policy-based scheduling mechanism to enhance system performance in grid. Also, we implement the proposed model and show that our policy-based scheduling mechanism can guarantee high system availability as well as support load balancing on an experimental basis.

QRP03-5: SLA-Constrained Resource Scheduling Policy for Group Peering in P2P Grid

Due to the dynamic nature of the grid computing and P2P in networks, the behavior of system might be very unpredictable in respect of reliability. Thus, we need well-organized system architecture to provide high system availability with resource scheduling scheme for P2P grid application. In this paper, we propose the SLA-constrained resource scheduling policy methodology in order to enhance system performance in P2P grid. The simulation results show that SCPA-based job scheduling can support the load balancing and guarantee the high system availability in system performance.

Policy-based resource management mechanism for dynamic Grid environments

As Grid networks grows, the complexity of resource management in Grid networks dramatically increases. To manage the Grid resources efficiently, policy based resource management system is suitable. However, the policy which made at a moment could not be appropriate to other time because the condition of the grid resources with time space changes. Thus, in this paper, we propose asynchronous policy-based Grid resource management mechanism for maintaining consistency between policies of Grid plane and network plane. Policy decision scheme for management cost and QoS guarantee make available resource quorum with Service Level Agreements (SLA), grid resource and network resource information in advance. If the available resource could not guarantee the userpsilas SLA in terms of completion time, service management suite divide the job into some sub-jobs to provide QoS service, and then process each sub-job. Through the adjustment scheme for maintaining consistency of policy, we can guarantee QoS service and stable service.

Adaptive management framework for scientific workflow applications

Nowadays, there are many scientific applications that consist of complex processes and services distributed across the different location. In order to meet user QoS, effective workflow scheduling is a key concern to be overcome for the execution of workflow applications in Grid. Thus, in this paper, we propose the adaptive resource management system that includes management functions for integrating workflow management with resource management system. In order to show the effectiveness of the proposed adaptive management framework, we implement OSICC (e-Organ Simulator Integrated Collaborative Computing) system that performs accurate and reliable simulation of the toxicity prediction for drug discovery. Furthermore, the experimental results showed that users could reduce simulation time in their drug discovery through the OSICC system.

Optimal Load Scheduling Mechanism and a Mobile-to-Grid Gateway Model for e-Health Service in Wireless Grid

The mission-critical application, in particular e-Health service such as ECG application and Carotid artery application, normally demands the highly sophisticated handling method and reliable processing scheme in wireless Grid environments. Due to the huge amounts of data set from the numerous wireless devices, we propose the refined procedures based on wireless Grid to guarantee the optimal processing time and reliability as prime factor of the system performance by using the proposed scheduling scheme. The simulation results show that proposed load scheduling can support the load balancing and guarantee the high system availability.

A Mobile-to-Grid Gateway Model and Load Scheduling Algorithm for e-Health Service in Wireless Grid

The mission-critical application, in particular e-health service such as ECG application and carotid artery application, normally demands the highly sophisticated handling method and reliable processing scheme in wireless grid environments. Due to the huge amounts of data set from the numerous wireless devices, we propose the refined procedures based on wireless grid to guarantee the optimal processing time and reliability as prime factor of the system performance by using the proposed scheduling scheme. The simulation results show that proposed load scheduling can support the load balancing and guarantee the high system availability.

A Cost-Optimized Redundancy Scheme of Group Peer in P2P Grid

Several aspects of todays grids are still based on centralized or hierarchical services. However, as grid sizes increase from tens to thousands of hosts, fault tolerance is becoming a key issue. In this paper, we propose a cost-optimized reliability using redundancy scheme in redundant group peers based P2P grid. Proposed finding optimum redundancy minimizes system cost and guarantees reliability which satisfies requirements of application under the dynamic behaviors in P2P grid environments

How Many SNPs Should Be Used for the Human Phylogeny of Highly Related Ethnicities? A Case of Pan Asian 63 Ethnicities

In planning a model-based phylogenic study for highly related ethnic data, the SNP marker number is an important factor to determine for relationship inferences. Genotype frequency data, utilizing a sub sampling method, from 63 Pan Asian ethnic groups was used for determining the minimum SNP number required to establish such relationships. Bootstrap random sub-samplings were done from 5.6K PASNPi SNP data. DA distance was calculated and neighbour-joining trees were drawn with every re-sampling data set. Consensus trees were made with the same 100 sub-samples and bootstrap proportions were calculated. The tree consistency to the one obtained from the whole marker set, improved with increasing marker numbers. The bootstrap proportions became reliable when more than 7,000 SNPs were used at a time. Within highly related ethnic groups, the minimum SNPs number for a robust neighbor-joining tree inference was about 7,000 for a 95% bootstrap support.

Economy-driven resource scheduling framework to accomplish the practical constraints in computational grid

Because of the dynamic nature and an instability of grid computing environment, system behavior in terms of QoS depends crucially on the selection of appropriate subset of the available resources. Thus, a well-defined system model that provides high system availability using a scheduling mechanism is required. In this paper, we propose the SLA based Policy Administrator (SPA) with QoS constraints with the resource scheduling scheme to find optimal resources for satisfying QoS requirements in terms of deadline as well as profit. Also, we demonstrate detailed workings of the proposed schemes with the SPA.