• 방송공학회논문지
• /
• 제24권6호
• /
• pp.974-982
• /
• 2019

게임 엔진을 기반으로 하는 실시간 360 VR(Virtual Reality) 스테레오 영상 획득 기술이 제안되었으나, 병목 현상이 발생하여 GPU(Graphics Processing Unit)의 성능을 충분히 활용하지 못 하고 있다. 본 논문에서는 기존에 제안된 실시간 360 VR 스테레오 영상 획득 기술의 병목 현상을 해결할 수 있도록 새로운 GPU 스케줄링 기법을 제안하고, 게임 엔진의 샘플 게임을 이용하여 제안하는 기법의 성능을 측정하였다. 측정 결과 기존에 제안된 기법보다 최대 약 70%의 성능 향상을 보였으며, GPU 자원이 좀더 균등하게 사용됨을 보였다.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제22권9호
• /
• pp.473-478
• /
• 2016

가상화 기술은 한정된 물리자원을 추상화하여 다수의 가상 자원 형태로 사용자에게 제공하는 기술로써 자원 활용률을 높이고 유동적으로 서비스를 제공할 수 있다는 장점이 있다. 하지만 한정된 물리자원을 다수의 가상머신이 공유하는 가상화 기술의 특성상, 자원 공유로 인한 성능 간섭 현상이 발생하는 문제가 있다. 이는 호스트 운영체제의 CPU 스케줄러가 가상머신에서 실행중인 프로세스의 특성을 고려하지 않고 스케줄링 하기 때문이다. 이러한 문제를 해결하기 위해 다양한 연구들이 진행되었지만 실제 근본적인 성능 간섭의 원인 분석에 대해서는 다루고 있지 않다. 본 논문에서는 KVM 기반 가상화 환경에서 가상머신의 성능 간섭의 원인을 분석하기 위해 다양한 시나리오에서의 프로파일링을 수행하고, 그 결과를 분석하여 CPU 스케줄링 관점에서 성능 간섭 현상의 원인과 그 해결 방안을 제시한다.

• 한국정보과학회논문지:시스템및이론
• /
• 제34권5_6호
• /
• pp.249-256
• /
• 2007

기존의 공정 스케줄링 방법들 중 GPS(generalized processor sharing)는 순간적 관점에서의 공정성을 추구하는 반면에, virtual clock은 장기적 관점에서의 공정성을 추구하는 특성을 지닌다. 이 논문에서는 이들의 차이가 과거의 서비스 정보를 추후의 스케줄링에 반영하는 정도에 있음에 주목하고, GPS와 virtual clock을 포괄하는 스펙트럼 형태의 스케줄링 모델을 제시한다. 이 모델에서 각 응용의 자원 획득 권한은 예치권한이라는 값으로 표현되는데, 예치권한은 각 응용별로 미리 정해진 고유한 비율로 계속 증가하며, 서비스를 받으면 소비된다. 소비되지 않고 누적된 예치권한은 과거에 서비스가 이루어지지 않은 정도를 표현하는 값이라고 볼 수 있으며, 이는 응용의 스케줄링 가능성을 높이므로 이후의 서비스 지연시간을 상대적으로 단축하는 효과를 낸다. 예치권한을 주기적으로 감쇄시키면 과거 정보의 반영 정도를 줄일 수 있으며, 이 때 그 감쇄 정도는 과거행태를 반영하는 정도를 의미한다. 과거의 정보를 전혀 반영하지 않을 경우 GPS의 특성을 나타내게 되며, 모두 반영할 경우 virtual clock의 특성을 보이게 된다. 이러한 스펙트럼 상에서는 평균지연시간과 장기적 공정성 사이에 절충 관계가 존재한다. 이 논문에서는 제시된 모델의 특성을 분석하고 실험을 통해 검증한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권9호
• /
• pp.4103-4121
• /
• 2018

Quality of Service (QoS) awareness is recognized as a key point for the success of Internet of Things (IOT).Realizing the full potential of the Internet of Things requires, a real-time task scheduling algorithm must be designed to meet the QoS need. In order to schedule tasks with diverse QoS requirements in cloud environment efficiently, we propose a task scheduling strategy based on dynamic priority and load balancing (DPLB) in this paper. The dynamic priority consisted of task value density and the urgency of the task execution, the priority is increased over time to insure that each task can be implemented in time. The scheduling decision variable is composed of time attractiveness considered earliest completion time (ECT) and load brightness considered load status information which by obtain from each virtual machine by topic-based publish/subscribe mechanism. Then sorting tasks by priority and first schedule the task with highest priority to the virtual machine in feasible VMs group which satisfy the QoS requirements of task with maximal. Finally, after this patch tasks are scheduled over, the task migration manager will start work to reduce the load balancing degree.The experimental results show that, compared with the Min-Min, Max-Min, WRR, GAs, and HBB-LB algorithm, the DPLB is more effective, it reduces the Makespan, balances the load of VMs, augments the success completed ratio of tasks before deadline and raises the profit of cloud service per second.

• 한국공작기계학회논문집
• /
• 제12권3호
• /
• pp.31-37
• /
• 2003

With the paradigm shifting from the principal of manufacturing efficiency to business globalism and rapid adaptation to its environments, more and more enterprises are being virtually organized as manufacturing network of different units in web. The formation of these enterprise called as Virtual Enterprise(VE) is becoming a growing trend as enterprises concentrating on core competence and economic benefit. 13us paper proposes multi-agent based task assignment system for VE, which attempts to address the selection of individually managed partners and the task assignment to them A case example is presented to illustrate how the proposed system can assign the task to partners.

• International Journal of Computer Science & Network Security
• /
• 제22권10호
• /
• pp.374-388
• /
• 2022

Cloud computing has been one of the most critical technology in the last few decades. It has been invented for several purposes as an example meeting the user requirements and is to satisfy the needs of the user in simple ways. Since cloud computing has been invented, it had followed the traditional approaches in elasticity, which is the key characteristic of cloud computing. Elasticity is that feature in cloud computing which is seeking to meet the needs of the user's with no interruption at run time. There are traditional approaches to do elasticity which have been conducted for several years and have been done with different modelling of mathematical. Even though mathematical modellings have done a forward step in meeting the user's needs, there is still a lack in the optimisation of elasticity. To optimise the elasticity in the cloud, it could be better to benefit of Machine Learning algorithms to predict upcoming workloads and assign them to the scheduling algorithm which would achieve an excellent provision of the cloud services and would improve the Quality of Service (QoS) and save power consumption. Therefore, this paper aims to investigate the use of machine learning techniques in order to predict the workload of Physical Hosts (PH) on the cloud and their energy consumption. The environment of the cloud will be the school of computing cloud testbed (SoC) which will host the experiments. The experiments will take on real applications with different behaviours, by changing workloads over time. The results of the experiments demonstrate that our machine learning techniques used in scheduling algorithm is able to predict the workload of physical hosts (CPU utilisation) and that would contribute to reducing power consumption by scheduling the upcoming virtual machines to the lowest CPU utilisation in the environment of physical hosts. Additionally, there are a number of tools, which are used and explored in this paper, such as the WEKA tool to train the real data to explore Machine learning algorithms and the Zabbix tool to monitor the power consumption before and after scheduling the virtual machines to physical hosts. Moreover, the methodology of the paper is the agile approach that helps us in achieving our solution and managing our paper effectively.

• 정보과학회 논문지
• /
• 제44권5호
• /
• pp.439-448
• /
• 2017

대규모 HPC 과학 응용의 워크로드가 전체 실행시간 동안 다양하게 변화하는 자원 요구사항을 갖게 되면서 특정 시점에 갑자기 요구사항이 증가하는(bursty) 형태가 되고 있다. 그러나 이러한 응용 워크로드를 고려하지 않고, 최대 자원 요구사항만을 반영한 가상 자원의 오버-프로비저닝은 과학 응용의 성능을 보장하지만 다른 응용이 사용할 수 없는 유휴 자원을 늘리는 문제로 남아있다. 본 논문에서는 OS-level 가상화 환경에서 응용의 자원 사용 패턴에 대한 프로파일링 데이터를 기반으로 메모리 자원 재구성 기법을 제안한다. 이는 유휴 상태의 메모리 자원을 신속하게 풀어주어 새로운 응용이 자원을 사용하여 수행할 수 있도록 한다. 본 연구에서는 경량화된 OS-level 가상화 시스템의 하나인 Docker에서 과학 워크플로우 응용을 이용하여 제안하는 알고리즘을 검증하였다. 실험을 통해 과학 응용을 실행하는 동안 컨테이너에 대한 메모리 할당 미세 조정이 전반적인 메모리 자원 활용을 향상시킬 수 있음을 보였다. 또한 응용의 메모리 사용 프로파일 데이터를 기반으로 하는 시뮬레이션 실험을 통해, 제안하는 동적 메모리 할당 기법을 사용하는 경우 대기 작업에 유휴상태의 메모리를 할당하여 전체 대기 작업의 수를 줄이고 시스템 작업 대기 시간이 줄어들었음을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권12호
• /
• pp.5694-5711
• /
• 2017

The technology of cloud computing is growing very quickly, thus it is required to manage the process of resource allocation. In this paper, load balancing algorithm based on honey bee behavior (LBA_HB) is proposed. Its main goal is distribute workload of multiple network links in the way that avoid underutilization and over utilization of the resources. This can be achieved by allocating the incoming task to a virtual machine (VM) which meets two conditions; number of tasks currently processing by this VM is less than number of tasks currently processing by other VMs and the deviation of this VM processing time from average processing time of all VMs is less than a threshold value. The proposed algorithm is compared with different scheduling algorithms; honey bee, ant colony, modified throttled and round robin algorithms. The results of experiments show the efficiency of the proposed algorithm in terms of execution time, response time, makespan, standard deviation of load, and degree of imbalance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5269-5286
• /
• 2018

Task scheduling is one of the key issues in improving system performance and optimizing resource management in cloud computing environment. In order to provide appropriate services for heterogeneous users, we propose a novel task scheduling strategy with service differentiation, in which the delay sensitive tasks are assigned to the rapid cloud with high-speed processing, whereas the fault sensitive tasks are assigned to the reliable cloud with service restoration. Considering that a user can receive service from either local SaaS (Software as a Service) servers or public IaaS (Infrastructure as a Service) cloud, we establish a hybrid queueing network based system model. With the assumption of Poisson arriving process, we analyze the system model in steady state. Moreover, we derive the performance measures in terms of average response time of the delay sensitive tasks and utilization of VMs (Virtual Machines) in reliable cloud. We provide experimental results to validate the proposed strategy and the system model. Furthermore, we investigate the Nash equilibrium behavior and the social optimization behavior of the delay sensitive tasks. Finally, we carry out an improved intelligent searching algorithm to obtain the optimal arrival rate of total tasks and present a pricing policy for the delay sensitive tasks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.2952-2971
• /
• 2016

Outsourcing jobs to a public cloud is a cost-effective way to address the problem of satisfying the peak resource demand when the local cloud has insufficient resources. In this paper, we studied the management of deadline-constrained bag-of-tasks jobs on hybrid clouds. We presented a binary nonlinear programming (BNP) problem to model the hybrid cloud management which minimizes rent cost from the public cloud while completes the jobs within their respective deadlines. To solve this BNP problem in polynomial time, we proposed a heuristic algorithm. The main idea is assigning the task closest to its deadline to current core until the core cannot finish any task within its deadline. When there is no available core, the algorithm adds an available physical machine (PM) with most capacity or rents a new virtual machine (VM) with highest cost-performance ratio. As there may be a workload imbalance between/among cores on a PM/VM after task assigning, we propose a task reassigning algorithm to balance them. Extensive experimental results show that our heuristic algorithm saves 16.2%-76% rent cost and improves 47.3%-182.8% resource utilizations satisfying deadline constraints, compared with first fit decreasing algorithm, and that our task reassigning algorithm improves the makespan of tasks up to 47.6%.