Rohaya Latip

Disaster Recovery in Cloud Computing: A Survey

Disaster recovery is a persistent problem in IT platforms. This problem is more crucial in cloud computing, because Cloud Service Providers (CSPs) have to provide the services to their customers even if the data center is down, due to a disaster. In the past few years, researchers have shown interest to disaster recovery using cloud computing, and a considerable amount of literature has been published in this area. However, to the best of our knowledge, there is a lack of precise survey for detailed analysis of cloud-based disaster recovery. To fill this gap, this paper provides an extensive survey of disaster recovery concepts and research in the cloud environments. We present different taxonomy of disaster recovery mechanisms, main challenges and proposed solutions. We also describe the cloud-based disaster recovery platforms and identify open issues related to disaster recovery.

Quorum-based Data Replication in Grid Environment

Replication is a useful technique for distributed database systems and can be implemented in a grid computation environment to provide a high availability, fault tolerant, and enhance the performance of the system. This paper discusses a new protocol named Diagonal Data Replication in 2D Mesh structure (DR2M) protocol where the performance addressed are data availability which is compared with the previous replication protocols, Read-One Write-All (ROWA), Voting (VT), Tree Quorum (TQ), Grid Configuration (GC), and Neighbor Replication on Grid (NRG). DR2M protocol is organized in a logical two dimensional mesh structure and by using quorums and voting techniques to improve the performance and availability of the replication protocol where it reduce the number of copies of data replication for read or write operations. The data file is copied at the selected node of the diagonal site in a quorum. The selection of a replica depends on the diagonal location of the structured two dimensional mesh quorum where ...

A particle swarm optimization and min---max-based workflow scheduling algorithm with QoS satisfaction for service-oriented grids

In service-orientated grids (SOG) environments, grid workflow schedulers play a critical role in providing quality-of-service (QoS) satisfaction for various end users (EUs) with diverse QoS objectives and optimization requirements. The EU requirements are not only many and conflicting, but also involve constraints of various degrees—loose, moderate or tight. However, most of the existing scheduling approaches violate EU constraints in tight situations and suffer inferior QoS optimization results. In this paper, a constraints-aware multi-QoS workflow scheduling strategy is proposed based on particle swarm optimization (PSO) and a proposed look-ahead heuristic (LAPSO) to improve performance in such situations. The algorithm selects the best scheduling solutions based on the proposed constraint-handling strategy. It hybridises PSO with a novel look-ahead mechanism based on a min–max heuristic, which deterministically improves the quality of the best solutions. Extensive simulation experiments have been carried out to evaluate the performance of the proposed approach. The simulation results show that the LAPSO algorithm guarantees satisfaction (0% violation) of the EU constraints even in tight situations. It also outperforms the comparison algorithm, with about 30% increase, in terms of cumulative QoS satisfaction of optimization requirements. In addition, the new scheme significantly reduces the CPU time by about 75% compared to the benchmark algorithm.

High Availability with Diagonal Replication in 2D Mesh (DR2M) Protocol for Grid Environment

Replication is a useful technique for distributed database systems and has been implemented in EU data grid and HEP in CERN for handling huge data access. Replica selection in their prototypes still can be enhanced to provide high availability, fault tolerant and low in communication cost. This paper introduces a new replica control protocol, named Diagonal Replication in 2D Mesh (DR2M) for grid environment and compares its performance with the previous protocols. The performance in this paper is data availability for read and write operation, which are compared to the Read-One Write-All (ROWA), Voting (VT), Tree Quorum (TQ), Grid Configuration (GC), Three Dimensional Grid Structure (TDGS), and Diagonal Replication on Grid (DRG). This paper discusses the protocol of replicating data for grid environment, putting the protocol in a logical 2D mesh structure by employing the quorums and voting techniques. The data file is copied in a selected replica from the diagonal sites in each quorum. The selection of a replica depends on the diagonal location of the structured 2D mesh network where the middle replica is selected because it is the shortest path to get a copy of the data from most of the direction in the quorum. The algorithm in this paper also calculates the optimized number of nodes to be grouped in each quorum and how many quorums are needed for the number of nodes, N in a network. DR2M protocol also ensures that the data for read and write operations are consistent, by ensuring the quorum must not have a nonempty intersection quorum. To evaluate the DR2M protocol, we developed a simulation model in Java. Our results prove that our protocol improves the performance of data availability compared to the previous data replication protocol, namely Read-One Write-All (ROWA), Voting (VT), Tree Quorum (TQ), Grid Configuration (GC), Three Dimensional Grid Structure (TDGS), and Diagonal Replication on Grid (DRG).

Point-to-Point Communication on Gigabit Ethernet and InfiniBand Networks

This paper presents the measurements of the MPI point-to-point communication performances on Razi and Haitham clusters by using SKaMPI, IMB and MPBench benchmark programs. These measurements were done on clusters with identical configurations in order to compare and analyze the MPI implementation performance over different interconnect technology; Gigabit Ethernet and InfiniBand. The comparison and analysis of the results from all benchmark programs used were then provided. It revealed that different MPI benchmark programs rendered different results for different interconnect. The results for both technologies were then compared to the experiment’s results that were done on cluster with Opteron quad-core processor. The comparison concluded that besides type of interconnect, the architecture of the clusters itself might also affected the results.

Workload utilization dissemination on grid resources for simulation environment

This paper discusses the workload utilization dissemination for grid computing. The CPU is a well-known resource item and it is an integral part in most literatures while other RIs may include memory, network and I/O overhead. The selection of resource variables and the number of RIs involved will result in different definitions of the workload. Various combination of computer RIs have been explored for studying the style of usage, techniques embedded and their capabilities. In contemplating the exploration, this study successfully describe the pattern of workload dissemination through the usage of the RIs and elicited the enhancement factors for systems performance. Among these factors are the manipulation of computer RIs, type of workload information with method of use, the workload dissemination direction along with implementation method and using certain algorithm to come out with new integrated scheduling with load balancing capability. A combination of these factors will help in developing an optimized scheduling or load balancing algorithm.

ADAPTIVE RESOURCE CONTROL MECHANISM THROUGH REPUTATION-BASED SCHEDULING IN HETEROGENEOUS DISTRIBUTED SYSTEMS

The service-oriented distributed systems such as Gr ids and Clouds are unified computing platform that connect and share heterogeneous resources including computation resource, storage resource, informatio n resource and knowledge resource. While these systems provide a vast amount of computing power their reliability are often hard to be guaranteed. It is due to the increased complexity of processing (e.g. , overhead, latency) that can indirectly affect the s ystem performance. In this study, we addressed the problem of dynamic control for resource management in distributed computing environment. Our dynamic resource control mechanism is designed based on rep utation-based scheduling that aims for sustainable resource sharing. Particularly, each computational resource in the environment has its own reputation value that calculated online by considering the computing capacity and availability. The degree of resource reputation significantly helps in scheduling decisi ons in terms of successful execution while adaptive ly monitoring resource availability. Results demonstra te that our resource control mechanism significantl y increases successful execution, while leading to ro bust resource management.

A SURVEY: PARTICLE SWARM OPTIMIZATION-BASED ALGORITHMS FOR GRID COMPUTING SCHEDULING SYSTEMS

Bio-inspired heuristics have been promising in solving complex scheduling optimization problems. Several researches have been conducted to tackle the problems of task scheduling for the heterogeneous and dynamic grid systems using different bio-inspired mechanisms such as Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO). PSO has been proven to have a relatively more promissing performance in dealing with most of the task scheduling challenges. However, to achieve optimum performance, new models and techniques for PSO need to be developed. This study surveys PSObased scheduling algorithms for Grid systems and presents a classification for the various approaches adopted. Meta task-based and workflow-based are the main categories explored. Each scheduling algorithm is described and discussed under the suitable category.

Splitting Strategies for Video Images Processing in Medical Data Grid

There has a lot of improvements had been done on the diagnostic tools. Due to these improvements, most of the medical images are now migrating from 1D (e.g. cardiograms and encephalograms) and 2D (e.g. x-rays) images to extended 3D (e.g. tomography) and eventually 4D (3 spatial dimension + time) image. This migration process has let the volume of medical image become larger and become difficult to store. To overcome this kind of problems, most of the hospitals start to integrate the grid technology into their medical storage system. In this paper, besides presented the current related research projects in the medical field, we also briefly discussed on the details of medical data grid, and the strategies and techniques used in the video and image processing area. From the discussion, based on the parameters of the initial delay and the deadline miss, we found that the Fibonacci-based splitting strategy is the most appropriate strategy for used in the video images processing.

A Comparative Study on Quorum-Based Replica Control Protocols for Grid Environment

Grid Computing handles huge amount of data which is stored in geographically distributed sites. It is a great challenge to ensure that data is managed, distributed and accessed efficiently in the distributed systems such as the data grid. To address the challenge, various techniques have been proposed in the literature. One of the widely used techniques is data replication since it offers high data availability, fault-tolerance and improve the performance of the system. In replication-based systems, replica control protocols are implemented for managing and accessing the data. In this paper, we present a comparison of various quorum-based replica control protocols that have been proposed in the distributed environment. This paper attempts to compare these replica control protocols based on the strengths, weaknesses and performance of the protocols.