• Journal of Korea Society of Digital Industry and Information Management
• /
• v.8 no.1
• /
• pp.99-105
• /
• 2012

Given the numbers of smartphones, tablets and other mobile devices shipped every day, more and more users are relying on the cloud as the main driver for satisfying their computing needs, whether it is data storage, applications or infrastructure. Mobile cloud computing is simply cloud computing in which at least some of the devices involved are mobile. Each node is owned by a different user and is likely to be mobile. Using mobile hardware for cloud computing has advantages over using traditional hardware. These advantage include computational access to multimedia and sensor data without the need for large network transfer, more efficient access to data stored on other mobile devices and distributed ownership and maintenance of hardware. It is important to predict job execution time in mobile cloud computing because there are many mobile nodes with different capabilities. This paper analyzes the job execution time for mobile cloud computing in terms of network environment and heterogeneous mobile nodes using a mathematical model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.670-686
• /
• 2017

MapReduce (MRV1), a popular programming model, proposed by Google, has been well used to process large datasets in Hadoop, an open source cloud platform. Its new version MapReduce 2.0 (MRV2) developed along with the emerging of Yarn has achieved obvious improvement over MRV1. However, MRV2 suffers from long finishing time on certain types of jobs. Speculative Execution (SE) has been presented as an approach to the problem above by backing up those delayed jobs from low-performance machines to higher ones. In this paper, an adaptive SE strategy (ASE) is presented in Hadoop-2.6.0. Experiment results have depicted that the ASE duplicates tasks according to real-time resources usage among work nodes in a cloud. In addition, the performance of MRV2 is largely improved using the ASE strategy on job execution time and resource consumption, whether in a multi-job environment.

• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.4
• /
• pp.293-304
• /
• 2005

In this paper, we research on job migration in a grid computing environment with cactus and MPICH-C2 based on Globus. Our concepts are to perform job migration by finding the site with plenty of computational resources that would decrease execution time in a grid computing environment. The Migration Manager recovers the job from the checkpointing files and restarts the job on the migrated site. To select a migrating site, the proposed method considers system's performance index, cpu's load, network traffic to send migration job tiles and the execution time predicted on a migration site. Then it selects a site with maximal performance gains. By selecting a site with minimum migration time and minimum execution time. this approach implements a more efficient grid computing environment. The proposed method Is proved by effectively decreasing total execution time at the $K\ast{Grid}$.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.4
• /
• pp.17-24
• /
• 2017

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.1
• /
• pp.147-153
• /
• 2016

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.1
• /
• pp.125-129
• /
• 2016

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.

• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.3
• /
• pp.127-137
• /
• 2010

As large-scale computational applications in various scientific domains have been utilized over many integrated sets of grid computing resources, the difficulty of their execution management and control has been increased. It is beneficial to refer job history generated from many application executions, in order to identify application‘s characteristics and to decide selection policies of grid resource meaningfully. In this paper, we apply a statistical technique, Plackett-Burman design with fold-over (PBDF), for analyzing grid environments and execution history of applications. PBDF design identifies main factors in grid environments and applications, ranks based on how much they affect to their execution time. The effective factors are used for selecting reference job profiles and then preferable resource based on the reference profiles is chosen. An application is performed on the selected resource and its execution result is added to job history. Factor's credit is adjusted according to the actual execution time. For a proof-of-concept, we analyzed job history from an aerospace research grid system to get characteristics of grid resource and applications. We built JARS algorithm and simulated the algorithm with the analyzed job history. The simulation result shows good reliability and considerable performance in grid environment with frequently crashed resources.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.1-8
• /
• 2010

In this paper, we propose a scheduling strategy for grid environment that reduces resource cost. This strategy considers resource cost and job failure rate to efficiently allocate local computing resources. The key idea of our strategy is that we use two-level scheduling using remote and local scheduler. The remote scheduler determines the expected total execution times of jobs using the current network and local system status maintained in its resource database and allocates jobs with minimum total execution time to local systems. The local scheduler recalculates the waiting time and execution time of allocated job and uses it to determine whether the job can be processed within the specified deadline. If it cannot finish in time, the job is migrated other local systems, through simulation, we show that it is more effective to reduce the resource cost than the previous Greedy strategy. We also show that the proposed strategy improves the performance compared to previous Greedy strategy.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.15 no.5
• /
• pp.511-518
• /
• 2015

As the cost-per-byte of SSDs dramatically decreases, the introduction of SSDs to Hadoop becomes an attractive choice for high performance data processing. In this paper the cost-per-performance of SSD-based Hadoop cluster (SSD-Hadoop) and HDD-based Hadoop cluster (HDD-Hadoop) are evaluated. For this, we propose a MapReduce performance model using queuing network to simulate the execution time of MapReduce job with varying cluster size. To achieve an accurate model, the execution time distribution of MapReduce job is carefully profiled. The developed model can precisely predict the execution time of MapReduce jobs with less than 7% difference for most cases. It is also found that SSD-Hadoop is 20% more cost efficient than HDD-Hadoop because SSD-Hadoop needs a smaller number of nodes than HDD-Hadoop to achieve a comparable performance, according to the results of simulation with varying the number of cluster nodes.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.5
• /
• pp.45-52
• /
• 2008

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-Performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.