• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10c
• /
• pp.304-306
• /
• 2002

이질적인 계산자원들로 구성된 환경에서 독립적인 작업들을 스케줄링하기 위한 최적의 방법을 찾는 것은 NP-Complete 문제로 알려져 있다 [4]. 현재까지 이 문제를 풀기 위한 다양한 휴리스틱 스케줄링 방법이 연구되어 왔다 [1, 8, 9, 10]. 작업의 선후 관계를 예측할 수 없는 상황에서는 동적 스케줄링 방법을 사용하며 동적 스케줄링 방법은 스케줄링 시기에 따라 온라인방식과 배치방식으로 나누어진다 [1, 12]. 본 논문에서는 새로운 스케줄링 알고리즘을 제안하였으며 제안된 스케줄링 알고리즘의 성능이 기존의 스케줄링 알고리즘의 성능보다 뛰어남을 실험을 통하여 보였다.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.6 no.6
• /
• pp.377-382
• /
• 2011

This paper describes the method of data synchronization among devices for personal cloud services. Existing data synchronization for mobile devices is based on a central server to mobile devices or a PC to a mobile device. However, the purpose of this paper is to share user data in heterogeneous environments, without depending on central server. This technology can be applied to synchronize personal data between a device and a personal cloud storage for personal cloud services. The ad hoc synchronization needs a sync agent service discovery module, a user authentication module, a network adapter, and an application data synchronization module. The method described in this paper is better than existing synchronization technology based on client-server in availability, performance, scalability quality attributes.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.233-240
• /
• 2021

We propose a novel, highly accurate approximate multiplier using different types of inexact 4-2 compressors. The importance of low hardware costs leads us to develop approximate multiplication for error-resilient applications. Several rules are developed when selecting a topology for designing the proposed multiplier. Our highly accurate multiplier design considers the different error characteristics of adopted compressors, which achieves a good error distribution, including a low relative error of 0.02% in the 8-bit multiplication. Our analysis shows that the proposed multiplier significantly reduces power consumption and area by 45% and 26%, compared with the exact multiplier. Notably, a trade-off relationship between error characteristics and hardware costs can be achieved when considering those of existing highly accurate approximate multipliers. In the image blending, edge detection and image sharpening applications, the proposed 8-bit approximate multiplier shows better performance in terms of image quality metrics compared with other highly accurate approximate multipliers.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.111-114
• /
• 2013

최근 고성능 컴퓨팅과 모바일 컴퓨팅에서 GPU 등의 성능가속기 사용이 증가함에 따라 성능가속기를 사용한 다양한 병렬처리 방법이 소개되고 있다. 그러나 성능 가속기를 처음 접하거나 성능가속기를 사용한 병렬처리 경험이 적은 사용자의 경우, 이러한 성능가속기를 이용하여 효과적인 병렬처리를 하는 것은 쉽지 않다. 본 논문에서는 성능가속기와 마이크로프로세서를 동시에 사용하여 단순히 성능가속기만을 사용한 병렬처리보다 효율적인 병렬처리 방법을 제안하고, 성능가속기만을 사용하여 얻은 성능과 제안한 방법의 성능을 비교한다. 실험결과, 제안방법은 순차처리와 비교하여 약 40배의 성능 향상을 얻을 수 있었고, 성능가속기만을 사용한 병렬처리 방법보다도 25%의 성능 향상이 가능함을 확인하였다.

• Journal of Internet Computing and Services
• /
• v.24 no.1
• /
• pp.17-26
• /
• 2023

A smart city is a massive internet of things (IoT) environment, where all terminal devices are connected to a network to create and share information. In accordance with massive IoT environments, millions of IoT devices are connected, and countless data are generated in real time. However, since heterogeneous IoT devices are used, collecting the logs for each IoT device is difficult. Due to these issues, when an IoT device is invaded or is engaged in malicious behavior, such as infection with malware, it is difficult to respond quickly, and additional damage may occur due to information leakage or stopping the IoT device. To solve this problem, in this paper, we propose identifying the attack technique used for initial access to IoT devices through MITRE ATT&CK, collect the logs that can be generated from the identified attack technique, and use them to identify the cause of malware infection.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.34-36
• /
• 2023

여러 인공지능 서비스의 보급은 초고성능 컴퓨팅 시스템 아키텍처의 변화를 야기하였고 다양한 계산 자원들의 활용이 모색되고 있다. 본 연구에서는 이러한 계산 자원들의 수용을 위해 범용적으로 사용되는 PCIe 버스를 기반으로 시스템 버스 확장 장치를 설계하고 구현하였다. PCIe 4.0 스위치를 기반으로 하는 확장 보드와 어댑터 카드를 개발하였고 GPU를 활용하여 실제 시스템으로의 활용 가능성을 검증하였다.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.126-133
• /
• 2023

Recent advancements in cloud service virtualization technologies have witnessed a shift from a Virtual Machine-centric approach to a container-centric paradigm, offering advantages such as faster deployment and enhanced portability. Container orchestration has emerged as a key technology for efficient management and scheduling of these containers. However, with the increasing complexity and diversity of heterogeneous workloads and service types, resource scheduling has become a challenging task. Various research endeavors are underway to address the challenges posed by diverse workloads and services. Yet, a systematic approach to container orchestration for effective cloud management has not been clearly defined. This paper proposes the DRA-Engine (Dynamic Resource Allocation Engine) for resource scheduling in container orchestration. The proposed engine comprises the Request Load Procedure, Required Resource Measurement Procedure, and Resource Provision Decision Procedure. Through these components, the DRA-Engine dynamically allocates resources according to the application's requirements, presenting a solution to the challenges of resource scheduling in container orchestration.

• The Journal of the Korea Contents Association
• /
• v.15 no.9
• /
• pp.21-28
• /
• 2015

In the semi-conductor process, a simulation process is performed to detect defects by analyzing the behavior of the impurity through the physical quantity calculation of the inner element. In order to perform the simulation, Finite-Difference Time-Domain(FDTD) algorithm is used. The improvement of semiconductor which is composed of nanoscale elements, the size of simulation is getting bigger. Problems that a processor such as CPU or GPU cannot perform the simulation due to the massive size of matrix or a computer consist of multiple processors cannot handle a massive FDTD may come up. For those problems, studies are performed with parallel/distributed computing. However, in the past, only single type of processor was used. In GPU's case, it performs fast, but at the same time, it has limited memory. On the other hand, in CPU, it performs slower than that of GPU. To solve the problem, we implemented a computing model that can handle any FDTD simulation regardless of size on the cluster which consist of heterogeneous processors. We tested the simulation on processors using MPI libraries which is based on 'point to point' communication and verified that it operates correctly regardless of the number of node and type. Also, we analyzed the performance by measuring the total execution time and specific time for the simulation on each test.

• Journal of KIISE:Software and Applications
• /
• v.35 no.12
• /
• pp.716-730
• /
• 2008

In Service-Oriented Architecture (SOA), service providers develop and deploy reusable services on the repositories, and service consumers utilize blackbox form of services through their interfaces. Services are also highly evolvable and often heterogeneous. Due to these characteristics of the service, it is hard to manage the faults if faults occur on the services. Autonomic Computing (AC) is a way of designing systems which can manage themselves without direct human intervention. Applying the key disciplines of AC to service management is appealing since key technical issues for service management can be effectively resolved by AC. In this paper, we present a theoretical model, Symptom-Cause-Actuator (SCA), to enable autonomous service fault management in SOA. We derive SCA model from our rigorous observation on how physicians treat patients. In this paper, we first define a five-phase computing model and meta-model of SCA. And, we define a schema of SCA profile, which contains instances of symptoms, causes, actuators and their dependency values in a machine readable form. Then, we present detailed algorithms for the five phases that are used to manage faults the services. To show the applicability of our approach, we demonstrate the result of our case study for the domain of 'Flight Ticket Management Services'.

• Journal of Internet Computing and Services
• /
• v.12 no.4
• /
• pp.1-14
• /
• 2011

This paper analyzes the fairness and efficiency of channel sharing when heterogeneous wireless networks that have different transmission power and/or coverage coexist with the contention-based channel access protocol. First, we show that the existing CSMA (carrier sensing multiple access) protocol, that is a prevailing contention-based mechanism, results in significant unfairness of channel access because of (1) the asymmetric capability of carrier sensing and (2) the blindness of binary exponential backoff and link adaptation mechanisms to the interference-driven transmission failures. Next, we derive the feasible region of carrier sensing thresholds that assures spatial reuse and fair channel sharing simultaneously. Moreover, we establish an analytical model for per-system throughput and investigate the effect of contention window size and transmission rate on the fairness and efficiency of channel sharing. Finally, we compare the performance of several approaches for fair channel sharing via simulations under various network configurations.