Daeyong Jung

An efficient checkpointing scheme using price history of spot instances in cloud computing environment

The cloud computing is a computing paradigm that users can rent computing resources from service providers as much as they require. A spot instance in cloud computing helps a user to utilize resources with less expensive cost, even if it is unreliable. When a user performs tasks with unreliable spot instances, failures inevitably lead to the delay of task completion time and cause a seriously deterioration in the QoS of users. Therefore, we propose a price history based checkpointing scheme to avoid the delay of task completion time. The proposed checkpointing scheme reduces the number of checkpoint trials and improves the performance of task execution. The simulation results show that our scheme outperforms the existing checkpointing schemes in terms of the reduction of both the number of checkpoint trials and total costs per spot instance for users bid.

A workflow scheduling technique using genetic algorithm in spot instance-based cloud

Cloud computing is a computing paradigm in which users can rent computing resources from service providers according to their requirements. A spot instance in cloud computing helps a user to obtain resources at a lower cost. However, a crucial weakness of spot instances is that the resources can be unreliable anytime due to the fluctuation of instance prices, resulting in increasing the failure time of users’ job. In this paper, we propose a Genetic Algorithm (GA)-based workflow scheduling scheme that can find the optimal task size of each instance in a spot instance-based cloud computing environment without increasing users’ budgets. Our scheme reduces total task execution time even if an out-of-bid situation occurs in an instance. The simulation results, based on a before-and-after GA comparison, reveal that our scheme achieves performance improvements in terms of reducing the task execution time on average by 7.06%. Additionally, the cost in our scheme is similar to that when GA is not applied. Therefore, our scheme can achieve better performance than the existing scheme, by optimizing the task size allocated to each available instance throughout the evolutionary process of GA.

An effective job replication technique based on reliability and performance in mobile grids

Recently, many studies have attempted to utilize mobile nodes as resources in mobile grids Due to their underlying restrictions such as intermittent communication disconnections, limited battery capacity, and so on, mobile nodes are less reliable than wired nodes for job processing Therefore, it is imperative to find an enhanced job scheduling method to provide stable job processing for mobile grids In this paper, we propose an efficient job scheduling method in mobile grids, which can determine the suitable number of replicas for a job based on resource (mobile node) information, node status, and access point information In our job scheduling method, mobile nodes are divided into node groups, and the number of subjobs assigned to each node group is derived from the reliability and performance of the node group Simulation results show that our scheduling algorithms can reduce the makespan of entire jobs in mobile grid environments compared with random-based job scheduling.

A Workflow Scheduling Technique for Task Distribution in Spot Instance-Based Cloud

The cloud computing is a computing paradigm that users can rent computing resources from service providers as much as they require. A spot instance in cloud computing helps a user to utilize resources with less expensive cost, even if it is unreliable. In this paper, we propose the workflow scheduling scheme that reduces the task waiting time when an instance occurs the out-of-bid situation. And, our scheme executes user’s job within selected instances and expands the suggested user budget. The simulation results reveal that, compared to various instance types, our scheme achieves performance improvements in terms of an average execution time of 66.86% over shortest execution time in each task time interval. And, the cost in our scheme is higher than an instance with low performance and is lower than an instance with high performance. Therefore, our scheme is difficult to optimize cost for task execution.

VM Migration for Fault Tolerance in Spot Instance Based Cloud Computing

The cloud computing is a computing paradigm that users can rent computing resources from service providers as much as they require. A spot instance in cloud computing helps a user to utilize resources with less expensive cost, even if it is unreliable. When a user performs tasks with unreliable spot instances, failures inevitably lead to the delay of task completion time and cause a seriously deterioration in the QoS of users. To solve the problem, we propose the VM migration scheme to reduce the job waiting time. And in this scheme we use our previously proposed checkpointing method. When a running instance occurs the out-of-bid situation (failure), the VM on the failed instance is to a new instance. Our proposed VM migration scheme reduces the rollback time and the task waiting time when an instance occur the out-of-bid situation. The simulation results show that our scheme achieves performance improvements in the task execution time of 68.94%, 68.61%, and 46.35% compared with the hour-boundary checkpointing scheme, the rising edge-driven checkpointing scheme, and our previously proposed checkpointing scheme., respectively Further, our scheme outperforms the existing schemes in terms of the reduction the total costs per spot instances for a user’s bid.

Estimated Interval-Based Checkpointing (EIC) on Spot Instances in Cloud Computing

In cloud computing, users can rent computing resources from service providers according to their demand. Spot instances are unreliable resources provided by cloud computing services at low monetary cost. When users perform tasks on spot instances, there is an inevitable risk of failures that causes the delay of task execution time, resulting in a serious deterioration of quality of service (QoS). To deal with the problem on spot instances, we propose an estimated interval-based checkpointing (EIC) using weighted moving average. Our scheme sets the thresholds of price and execution time based on history. Whenever the actual price and the execution time cross over the thresholds, the system saves the state of spot instances. The Bollinger Bands is adopted to inform the ranges of estimated cost and execution time for users discretion. The simulation results reveal that, compared to the HBC and REC, the EIC reduces the number of checkpoints and the rollback time. Consequently, the task execution time is decreased with EIC by HBC and REC. The EIC also provides the benefit of the cost reduction by HBC and REC, on average. We also found that the actual cost and execution time fall within the estimated ranges suggested by the Bollinger Bands.

Task Balanced Workflow Scheduling Technique considering Task Processing Rate in Spot Market

Recently, the cloud computing is a computing paradigm that constitutes an advanced computing environment that evolved from the distributed computing. And the cloud computing provides acquired computing resources in a pay-as-you-go manner. For example, Amazon EC2 offers the Infrastructure-as-a-Service (IaaS) instances in three different ways with different price, reliability, and various performances of instances. Our study is based on the environment using spot instances. Spot instances can significantly decrease costs compared to reserved and on-demand instances. However, spot instances give a more unreliable environment than other instances. In this paper, we propose the workflow scheduling scheme that reduces the out-of-bid situation. Consequently, the total task completion time is decreased. The simulation results reveal that, compared to various instance types, our scheme achieves performance improvements in terms of an average combined metric of 12.76% over workflow scheme without considering the processing rate. However, the cost in our scheme is higher than an instance with low performance and is lower than an instance with high performance.

An Estimation-Based Task Load Balancing Scheduling in Spot Clouds

Cloud computing is a computing paradigm in which users can rent computing resources from service providers according to their requirements. Cloud computing based on the spot market helps a user to obtain resources at a lower cost. However, these resources may be unreliable. In this paper, we propose an estimation-based distributed task workflow scheduling scheme that reduces the estimated generation compared to Genetic Algorithm (GA). Moreover, our scheme executes a user’s job within selected instances and stretches the user’s cost. The simulation results, based on a before-and-after estimation comparison, reveal that the task size is determined based on the performance of each instance and the task is distributed among the different instances. Therefore, our proposed estimation-based task load balancing scheduling technique achieves the task load balancing according to the performance of instances.

A Workflow Scheduling Technique to Consider Task Processing Rate in Spot Instance-Based Cloud

Cloud computing is a computing paradigm in which users can rent computing resources from service providers according to their requirements. A spot instance in cloud computing helps a user to obtain resources at a lower cost, but it may be unreliable. In this paper, we propose a workflow scheduling scheme to consider task processing rate. This scheme reduces the task waiting time and the rollback time when an out-of-bid situation occurs in an instance.

A Study on Performance Comparison of Cloud Architectures Using Nested Virtualization

In recent years, cloud computing has become a significant technology trend because of various advantages including cost savings, flexibility, high availability, and scalability using virtualization technology. However, one of the concerns for using cloud computing is security. In fact, there are multiple attack surfaces in virtualized environments. In this paper, we build a fault tolerant cloud architecture using nested virtualization. With our constructed cloud architecture, we argue that a malicious virtual machine cannot subvert the whole virtual machines on the physical host machine. To support this, we compare the performance of two types of virtual machines (i.e., nested virtual machines and regular virtual machines). Results provide encouraging support for the validity of our cloud architecture with negligible performance degradation while having fault tolerance.