[Show abstract][Hide abstract] ABSTRACT: The cloud environment makes it possible to analyze large data sets in a scalable computing infrastructure. In the bioinformatics field, the applications are composed of the complex workflow tasks, which require huge data storage as well as a computing-intensive parallel workload. Many approaches have been introduced in distributed solutions. However, they focus on static resource provisioning with a batch-processing scheme in a local computing farm and data storage. In the case of a large-scale workflow system, it is inevitable and valuable to outsource the entire or a part of their tasks to public clouds for reducing resource costs. The problems, however, occurred at the transfer time for huge dataset as well as there being an unbalanced completion time of different problem sizes. In this paper, we propose an adaptive resource-provisioning scheme that includes run-time data distribution and collection services for hiding the data transfer time. The proposed adaptive resource-provisioning scheme optimizes the allocation ratio of computing elements to the different datasets in order to minimize the total makespan under resource constraints. We conducted the experiments with a well-known sequence alignment algorithm and the results showed that the proposed scheme is efficient for the cloud environment.
Preview · Article · Dec 2012 · Journal of Information Processing Systems
[Show abstract][Hide abstract] ABSTRACT: For widely distributed data analysis applications over that run the Internet, both the instability of the data transfer time and the dynamics of data processing rate require a more sophisticated data provisioning scheme to maximize parallel efficiency, in particular, under conditions in which real-time and limited data buffer (storage) constraints are given. In this letter, we propose a synchronized data provisioning scheme that implicitly avoids the data buffer overflow as well as explicitly controls the data buffer underflow by optimally adjusting the buffer resilience. In order to guarantee the designated quality of service, we further exploit an adaptive buffer resilience control algorithm based on sample path analysis of the state of the data buffer and the demand queue. The simulation results show that the proposed scheme is suitably efficient to apply to an environment that can not postulate the stochastic characteristics of the data transfer time and data processing rate.
[Show abstract][Hide abstract] ABSTRACT: Scientific and engineering problems often require a large amount of computing resources to conduct large-scale simulations with a wide range of input parameters. This paper describes an application parameter meta service and agent-controlled loop scheduling for executing multiple parameter sweep applications in a distributed environment. The proposed system provides a flexible and practical environment for generating a large set of independent jobs and for scheduling them over distributed resources on the grid. Using the proposed environment, we conducted a protein folding experiment that had about four thousands initial data on the pacific rim applications and grid middleware assembly (PRAGMA) testbed. The folding optimization provided 3-dimensional protein atomic structures.
Full-text · Article · Nov 2009 · Journal- Korean Physical Society
[Show abstract][Hide abstract] ABSTRACT: Due to the dynamic nature and uncertainty of grid computing, system reliability can become very unpredictable. Thus, 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.
No preview · Article · Dec 2008 · IEICE Transactions on Communications
[Show abstract][Hide abstract] ABSTRACT: To utilize an e-Science application service which is built on a grid computing environment efficiently, it is vital to develop technologies for integrating each e- Science application service into a workflow and to execute the workflow in a cooperative environment. In this paper, the required workflow in e-Science is analyzed using a conceptual scenario and categorized into three types of workflows: simulations, experiments, and collaborative work. Additionally, the workflow of the aerospace research community, an e-Science environment is shown. The necessary applications and functionalities for the required workflow are then proposed.
[Show abstract][Hide abstract] ABSTRACT: In e-science environments, scientists need to execute a scientific application with various parameters multiple times to simulate and experiment complicated problems on the grid. For this, they should write every single job description with distinct parameters even if this is a troublesome task. To provide the flexibility and adaptability for parameter study, we propose an application parameter description language (APDL) and a service oriented parameter study scheme, called a parametric study service (PSS), for parameterized simulations on the grid. The APDL extends the job submission description language (JSDL) to generate parameters for multiple jobs. The proposed PSS provides a unified interface to submit jobs into various middleware platforms such as gLite, Globus, etc. The problem solving environment (PSE) assists a parameter study for their applications and every research fields tend to construct individual own PSE., The proposed PSS can be easily adapted into the specific PSE because of being implemented as Web services. In practice, we apply the APDL and the PSS into aerospace research PSE which carry out the three-dimensional turbulent analysis for compressible flow.
[Show abstract][Hide abstract] ABSTRACT: This paper describes a parametric study service (PSS) that enables parameterized simulations on the grid. Scientific and engineering problems often require the large amount of computing resources to conduct a large-scale simulations with a wide range of input parameters and files, namely parametric study. Conducting parametric studies on the grid is a great challenge because it involves the allocation, scheduling and monitoring of jobs over the highly heterogeneous, distributed and failure-prone grid environments. The system exploits the grid computing technology to establish an integrated Web service environment which supports distributed computational simulations and remote executions. The proposed system provides an easy-to-use parametric study service that prevents users from being exposed to the complexity of the underlying grid. To demonstrate the usefulness of our PSS system, we have carried out thousands of protein folding simulations in the grid testbed, namely the KMI testbed, which consists of Linux clusters distributed across the nation.
[Show abstract][Hide abstract] ABSTRACT: Recently, there are many paradigms for supporting convergence for both networks and computing. Grid computing offers high levels of computational, storage and network capacity on the wide area environment. Optic based network is a good candidate for supporting the grid application. In this paper, we proposed a novel contention resolution scheme which is differentiated by specific application and their job type for supporting grid services in the OBS network
[Show abstract][Hide abstract] ABSTRACT: In wireless Grid computing environment, end-to-end Quality of Service (QoS) could be very complex, and this highlights the
increasing requirement for the management of QoS itself. Policy quorum-based management offers a more flexible, customizable
management solution that allows controlled elements to be configured or scheduled on the fly, for a specific requirement tailored
for a customer. This paper presents a QoS guaranteed management scheme, Policy Quorum Resource Management (PQRM), which facilitates
the reliable scheduling of the wireless Grid system with dynamic resource management architecture aimed at fulfilling the
QoS requirements. Experimental results show the proposed PQRM with resource reconfiguration scheme improves both performance
and stability, which is suitable for a wireless Grid services.
[Show abstract][Hide abstract] ABSTRACT: Data grids provide researching groups with data management services such as Grid-Ftp, Reliable File Transfer Service (RFP),
and Replica Location Service (RLS). Especially, Replica Location Service could improve data availability by storing multiple
data to the distributed location. In this paper, we propose Tree-based Replica Location Scheme (TRLS) that could support users
to decide optimally the location of multiple replicas as well as the number of replicas. In addition to, we could decide the
number of replicas to satisfy user’s requirements by minimizing cost
[Show abstract][Hide abstract] ABSTRACT: Quality of Service (QoS)-constrained policy has an advan- tage to guarantee QoS requirements requested by users. Quorum systems can ensure the consistency and availability of replicated data despite the benign failure of data repositories. We propose a Quorum based resource management scheme, which includes a system resource and network re- source, both of which can satisfy the requirements of application QoS. We also propose the resource reconfiguration algorithm based on tem- poral execution time estimation method. Resource reconfiguration per- forms the reshuffling of the current available resource set for maintaining the quality level of the resources. We evaluate the effectiveness of re- source reconfiguration mechanism in a Heart Hemodynamics analysis. Our approach increases the stability of execution environment as well as decreases the completion time compared to the method that does not adopt the proposed reconfiguration scheme.
[Show abstract][Hide abstract] ABSTRACT: Data grids provide researching groups with data management services such as Grid-Ftp, Reliable File Transfer Service (RFP), and Replica Location Service (RLS). Especially, Replica Location Service could improve data availability by storing multiple data to the distributed location. However, there is no economic scheme to decide the location of multiple replicas. In this paper, we propose Tree-based Replica Location Scheme (TRLS) that could support users to decide optimally the location of multiple replicas as well as the number of replicas. By TRLS, we could minimize the replication cost since it supports us to find an optimal location, satisfying both storage cost and communication cost at the same time. In addition to, we could decide the number of replicas to satisfy user's requirements. By minimizing cost, we make an economic decision, and we expect our proposed scheme could be regarded as a basic economic model.