• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.105-107
• /
• 2014

대량의 정보를 효과적으로 처리하기 위한 기술로 CPU 뿐만 아니라 iSSD 와 GPGPU 를 개별적으로 이용하는 연구가 진행되고 있다. 본 논문에서는 더 나아가 CPU, iSSD 와 GPU 를 협업시켜 프로그램 수행 성능을 향상시키는 방법을 연구한다. 이러한 이질 시스템의 협업을 위해 이질 스케줄링 알고리즘을 적용하고, 스케줄링 알고리즘에 따른 협업 시스템의 성능을 분석한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.52-57
• /
• 2007

한정된 자원을 효율적으로 사용해야하는 이동통신망에서 멀티미디어 서비스 요구에 따른 무선 트래픽 채널을 할당하는 기법은 무선이라는 특수 환경으로 인해 제약을 받을 수밖에 없다. 이동망의 기지국의 경우 여러 무선 가입자 보드로부터 요구되는 서비스별 트래픽요구에 대한 채널 할당과 이에 대한 메인보드에서 처리해야 하는 작업 스케줄링은 무선과 CPU라는 서로 다른 환경을 잘 매핑하는 과제를 안고 있다. 본 논문에서는 음성과 데이터 호를 동시에 서비스하는 셀룰러 시스템에서 멀티미디어 서비스 트래픽 특성을 고려한 주파수할당과 작업 스케줄링이라는 두 가지 요소를 접목할 때 인공지능알고리즘인 유전자알고리즘을 이용하는 방법과 이에 적합한 작업 스케줄링 방식을 제안한다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.18 no.2
• /
• pp.141-147
• /
• 1992

This article evaluates the efficiency of three branch-and-bound heuristics for a job scheduling problem that minimizes the sum of absolute deviations of completion times from a common due date. To improve the performance of the branch-and-bound procedure, Algorithm SA is presented for the initial feasible schedule and three heuristics : breadth-first, depth-first and best-first search are investigated depending on the candidate selection procedure. For the three heuristics the CPU time, memory space, and the number of nodes generated are computed and tested with nine small examples (6 ${\leq}$ n ${\leq}$ 4). Medium sized random problems (10 ${\leq}$ n ${\leq}$ 30) are also generated and examined. The computational results are compared and discussed for the three heuristics.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.3
• /
• pp.193-204
• /
• 2006

In most existing distributed Web systems, incoming requests are distributed to servers via Domain Name System (DNS). Although such systems are simple to implement, the address caching mechanism easily results in load unbalancing among servers. Moreover, modification of the DNS is necessary to load considering the server's state. In this paper, we propose a new dynamic load balancing method using dynamic DNS update and round-robin mechanism. The proposed method performs effective load balancing without modification of the DNS. In this method, a server can dynamically be added to or removed from the DNS list according to the server's load. By removing the overloaded server from the DNS list, the response time becomes faster. For dynamic scheduling, we propose a scheduling algorithm that considers the CPU, memory, and network usage. We can select a scheduling policy based on resources usage. The proposed system can easily be managed by a GUI-based management tool. Experiments show that modules implemented in this paper have low impact on the proposed system. Furthermore, experiments show that both the response time and the file transfer rate of the proposed system are faster than those of a pure Round-Robin DNS.

• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.193-198
• /
• 2022

Face Detection is one of the most important aspects of image processing, it considers a time-consuming problem in real-time applications such as surveillance systems, face recognition systems, attendance system and many. At present, commodity hardware is getting more and more heterogeneity in terms of architectures such as GPU and MIC co-processors. Utilizing those co-processors along with the existing traditional CPUs gives the algorithm a better chance to make use of both architectures to achieve faster implementations. This paper presents a hybrid implementation of the face detection based on the local binary pattern (LBP) algorithm that is deployed on both traditional CPU and MIC co-processor to enhance the speed of the LBP algorithm. The experimental results show that the proposed implementation achieved improvement in speed by 3X when compared to a single architecture individually.

• Journal of Internet of Things and Convergence
• /
• v.3 no.1
• /
• pp.21-24
• /
• 2017

Modern computers have to deal with multiple processes. The operating system uses scheduling to handle many processes with a small number of CPUs. Types of scheduling include FCFS, SJF, and RR. Of these, RR shall determine the maximum allocation time. In this paper, we analyzed the GLM algorithm for specific samples to find the optimal maximum allocation time. This analysis method allows us to specify the maximum allocation time according to the desired conditions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.6
• /
• pp.1892-1912
• /
• 2022

With the massive demand and growth of cloud computing, virtualization plays an important role in providing services to end-users efficiently. However, with the increase in services over Cloud Computing, it is becoming more challenging to manage and run multiple Virtual Machines (VMs) in Cloud Computing because of excessive power consumption. It is thus important to overcome these challenges by adopting an efficient technique to manage and monitor the status of VMs in a cloud environment. Reduction of power/energy consumption can be done by managing VMs more effectively in the datacenters of the cloud environment by switching between the active and inactive states of a VM. As a result, energy consumption reduces carbon emissions, leading to green cloud computing. The proposed Efficient Dynamic VM Scheduling approach minimizes Service Level Agreement (SLA) violations and manages VM migration by lowering the energy consumption effectively along with the balanced load. In the proposed work, VM Scheduling for Efficient Dynamically Migrated VM (VMS-EDMVM) approach first detects the over-utilized host using the Modified Weighted Linear Regression (MWLR) algorithm and along with the dynamic utilization model for an underutilized host. Maximum Power Reduction and Reduced Time (MPRRT) approach has been developed for the VM selection followed by a two-phase Best-Fit CPU, BW (BFCB) VM Scheduling mechanism which is simulated in CloudSim based on the adaptive utilization threshold base. The proposed work achieved a Power consumption of 108.45 kWh, and the total SLA violation was 0.1%. The VM migration count was reduced to 2,202 times, revealing better performance as compared to other methods mentioned in this paper.

• Journal of KIISE:Software and Applications
• /
• v.31 no.2
• /
• pp.164-172
• /
• 2004

For a local search algorithm to find a bettor quality solution it is required to generate and evaluate a sufficiently large number of candidate solutions as neighbors at each iteration, demanding quite an amount of CPU time. This paper presents a method of selectively generating only good-looking candidate neighbors, so that the number of neighbors can be kept low to improve the efficiency of search. In our method, a newly generated candidate solution is probabilistically selected to become a neighbor based on the quality estimation determined heuristically by a very simple evaluation of the generated candidate. Experimental results on the problem of load balancing for production scheduling have shown that our candidate selection method outperforms other random or greedy selection methods in terms of solution quality given the same amount of CPU time.

• Journal of IKEEE
• /
• v.21 no.1
• /
• pp.92-95
• /
• 2017

In this paper, we proposed a way to improve function of small scale GP-GPU. Instead of using superscalar which increase scheduling-complexity, we suggested the application of simple core to maximize GP-GPU performance. Our studies also demonstrated that simplified Stream Processor is one of the way to achieve functional improvement in GP-GPU. In addition, we found that developing of optimal thread-assigning method in Warp Scheduler for specific application improves functional performance of GP-GPU. For examination of GP-GPU functional performance, we suggested the thread-assigning way which coordinated with Deep-Learning system; a part of Neural Network. As a result, we found that functional index in algorithm of Neural Network was increased to 90%, 98% compared with Intel CPU and ARM cortex-A15 4 core respectively.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.07a
• /
• pp.205-206
• /
• 2018

본 논문에서는 IoT 환경의 무선 센서 네트워크 시스템 상의 효율적인 패킷 전달을 위해 큐러닝(Q-learning)에 기반한 다중 대기열 동적 스케쥴링 기법을 제안한다. 이 정책은 다중 대기열(Multiple queue)의 각 큐가 요구하는 딜레이 조건에 맞춰 최대한 패킷 처리를 미룸으로써 효율적으로 CPU자원을 분배한다. 또한 각 노드들의 상태를 큐러닝(Q-learning)을 통해 지속적으로 상태를 파악하여 기아상태(Starvation)를 방지한다. 제안하는 기법은 무선 센서 네트워크 상의 가변적이고 예측 불가능한 환경에 대한 사전지식이 없이도 요구하는 서비스의 질(Quality of service)를 만족할 수 있도록 한다. 본 논문에서는 모의실험을 통해 기존의 학습 기반 패킷 스케쥴링 알고리즘과 비교하여 제안하는 스케쥴링 기법이 복잡한 요구조건에 따라 유연하고 공정한 서비스를 제공함에 있어 우수함을 증명하였다.