• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.509-512
• /
• 2002

VIA 는 클러스터 또는 시스템 영역 네트워크를 위한 표준화된 사용자수준 통신 아키텍쳐이고, GNBD 는 LINUX 클러스터에서 IP 네트워크 설비를 기반으로 GFS 공유 파일 시스템을 설치할 때 사용하는 네트워크 블록 디바이스이다. GNBD 는 TCP/IP 상의 소켓을 기반으로 구현되어 있기 때문에, VIA 를 사용하는 클러스터이더라도 VIA 하드웨어 상에서 TCP/IP 소켓을 통하여 GNBD 를 작동시킨다. VIA 와 같이 물리적 연결이 신뢰성이 높고 높은 수준의 기능을 제공하는 경우는 같은 클러스터 안에서 TCP/IP 프로토콜 스택을 사용할 필요가 없다. 그래서 우리는 VIA 를 이용하지만 TCP/IP를 사용하지 않는 GNBD/VIA를 구현하였고, 동일한 VIA 하드웨어를 사용하면서 TCP/IP 모듈을 이용하는 GNBD 보다 파일시스템의 읽기(쓰기) 성능이 약 20%(30%) 향상된다는 것을 확인하였다. 본 논문에서는 VIA상에서 동작하는 GNBD/VIA의 성능 측정값과 그 위에 설치된 파일시스템의 을 보여주고, 그 결과를 상세히 분석하여 GNBD/VIA 상에 설치된 파일 시스템이 발휘할 수 있는 성능의 한계를 제시한다. 제시하는 한계치는 GNBD/VIA 뿐만 아니라 TCP/IP 상의 소켓을 사용하는 GNBD에도 적용할 수 있다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.5
• /
• pp.199-202
• /
• 2013

In the cluster mechanism of the wireless sensor network, the network lifetime is affected by how cluster heads are selected. One of the representative clustering mechanisms, the low-energy adaptive clustering hierarchy (LEACH), selects cluster heads periodically, resulting in high energy consumption in cluster reconstruction. On the other hand, the adaptive clustering algorithm via waiting timer (ACAWT) proposes a non-periodic re-clustering mechanism that reconstructs clusters if the remaining energy level of a cluster head reaches a given threshold. In this paper, we propose a re-clustering mechanism that uses multiple remaining node energy levels and does re-clustering when the remaining energy level of a cluster head reaches one level lower. Also, in determining cluster heads, both of the number of neighbor nodes and the remaining energy level are considered so that cluster heads can be more evenly placed. From the simulations based on the Qualnet simulator, we validate that our proposed mechanism outperforms ACAWT in terms of the network lifetime.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.2
• /
• pp.221-233
• /
• 2008

In general, we use a Parallel processing computer manufactured specially for the purpose of parallel processing to do high performance computings. But we can depoly and use a PC cluster composed of several common PCs instead of the very expensive parallel processing computer. A common way to get a PC cluster is to adopt the star topology network connected by a switch hub. But in this paper, we grope efficient implementations of hypercube networks based on TCP/IP to connect 8 PCs directly for more useful parallel processing environment, and make evaluations on functionality and efficiency of them using ping, netperf, MPICH. The two proposed methods of implementation are IP configuration based on link and IP configuration based on node. The results of comparison between them show that there is not obvious difference in performance but the latter is more efficient in simplicity of routing table. For verification of functionality, we compare the parallel processing results of an application in them with the same in a star network based PC cluster. These results also show that the proposed hypercube networks support a perfect parallel processing environment respectively.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.792-795
• /
• 2012

애드 혹 네트워크(MANET: Mobile Ad hoc NETworks)는 기본적인 내부구조(infrastructure) 없이 노드들만으로 네트워크 망을 구성한다. 경로 탐색 정책으로 리액티브(reactive) 방식과 프로액티브(proactive) 방식이 있는데, 전통적으로 리액티브 방식의 성능이 더 좋은 것으로 평가된다. 그리고 두가지 방식의 장점을 취합한 하이브리드(hybrid) 방식의 클러스터 토폴로지(cluster topology) 도입에 관한 연구가 이루어지고 있다. 그 중, HCR(Hybrid Cluster Routing)이 제안되었는데, 이는 프로액티브 방식에 보다 중심을 둔 기법이다. HCR 은 리액티브 방식 경로 탐색 방법인 플라딩(flooding)의 탐색 지역을 한정된 범위로 제한할 수 있으나, 프로액티브 방식의 전체 네트워크 구성 정보 유지에 따른 막대한 오버헤드를 발생한다. 본 논문에서는 이러한 오버헤드를 줄이기 위해, 클러스터 내부 경로 탐색 기법인 MICF(Maginot path based Intra Cluster Flooding)를 제안한다. MICF 는 HCR 을 개선한 FSRS(First Search and Reverse Setting) 기반의 기법으로서, 클러스터 내부의 마지노 패스(maginot path)를 기준으로 경로 탐색 지역을 제한한다. MICF 는 게이트웨이(gateway) 간 최단 거리가 항상 클러스터 헤드(cluster head)를 중점으로 원의 내각 지역에 존재함을 바탕으로 하며, 최단 경로의 보장과 플라딩 지역 제한을 동시에 만족한다. 실험 결과, MICF 는 FSRS 기반의 기존 클러스터 내부 플라딩 방식보다 총 에너지의 7.79%만큼 더 에너지를 보존하였다. 결론적으로, MICF 역시 기존의 방식보다 에너지를 더 효율적으로 사용할 수 있으며, 마지노패스 설정과 이를 기반으로 한 제어 과정에 추가적인 오버헤드가 발생하지 않는다. 그리고 플라딩 면적이 작을수록 오버헤드가 줄어들게 됨을 알 수 있다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.6
• /
• pp.103-108
• /
• 2015

Maintaining efficient energy consumption and elongating network lifetime are the key issues in wireless sensor networks. Existing routing protocols usually select the cluster heads based on the proximity to the sensor nodes. In this case the cluster heads can be placed farther to the base station, than the distance between the sensor nodes and the base station, which yields inefficient energy consumption. In this work we propose a novel algorithm that select the nodes in a cluster and the cluster heads based on the locations of related nodes. We verify that the proposed algorithm gives better performance in terms of network life time than existing solutions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.11A
• /
• pp.1164-1172
• /
• 2007

To solve scalability problem in the ZigBee Network, this paper presents a new mesh routing protocol for ZigBee, called ZigBee Cluster Label (ZiCL). ZiCL divides the ZigBee network into one or more logical clusters and then assigns a unique Cluster Label to each cluster so that it discovers a route of a destination node based on Cluster Label. When a node collects new Cluster Label information of a destination node according to discovery based on Cluster Label, ZiCL encourages nodes with the same Cluster Label to share the information. Consequen tly, it contributes on reducing numerical potential route discoveries and improving network performances such as routing overhead, end-to-end delay, and packet delivery ratio. Simulation results using NS-2 show ZiCL performs well.

• Journal of IKEEE
• /
• v.11 no.4
• /
• pp.205-212
• /
• 2007

The ubiquitous sensor network consists of micro sensors with wireless communication capabilities. Compared to wired communication, wireless communication is more subject to eavesdropping as well as data variation and manipulation. Accordingly, there must be efforts to secure the information delivered over the sensor network. Providing security to the sensor network, however, requires additional energy consumption, which is an important issue since energy transformation is difficult to implement in a sensor network. This paper proposes a routing mechanism based on the energy-efficient cluster that features security functions capable of safely processing the data acquired from the sensor network. The proposed algorithm reduces energy consumption by fixing the clusters formed at the initial stage and using the pre-distribution scheme so that the cluster and node keys generated and exchanged at the initial stage are not re-generated or re-exchanged. Simulation experiments confirmed that the proposed approach reduces energy consumption compared to implementing security measures to the conventional cluster-based routing mechanism.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.2
• /
• pp.1-10
• /
• 2009

Energy efficient operations are essential to increase the life time of wireless sensor network. A cluster-based protocol is the most common approach to preserve energy during a data aggregation. This paper deals with an energy awareness and autonomous clustering method based on time delay. This method consists of three stages. In the first phase, Candidate Cluster Headers(CCHs) are selected based on a time delay which reflects the remaining energy of a node, with considering coverage efficiency of a cluster. Then, time delay is again applied to declare Cluster Headers(CHs) out of the CCHs. In the last phase, the issue on an orphan node which is not included into a cluster is resolved. The simulation results show that the proposed method increases the life time of the network around triple times longer than LEACH(Low Energy Adaptive Cluster Hierarchy). Moreover, the cluster header frequency is less diverse, and the energy on cluster heads is less spent.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.4
• /
• pp.1001-1008
• /
• 2015

Mobile Ad hoc Network(MANET) consists of a node that has mobility. In MANET, the node has routing, the node builds a network of their own, no dependent infrastructure. Topology are exchanged due to node mobility in MANET. For reducing the change of topology, hierarchical network algorithm has been investigated. In hierarchical network, cluster member node communicates through cluster head node. When the load-balancing of cluster head node is exceed, assigned cluster member node can't communicate with base station. To solve this problem, we proposed Load Tolerance algorithm. The proposed algorithm, when cluster member node can't send a message by cluster head node that exceed load-balancing, then the cluster member node sends a message by selected load tolerance node. Through a simulation, the proposed algorithm improves packet delivery ratio in cluster routing.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.325-327
• /
• 2005

이동 애드혹 네트워크에서 트리와 메쉬 기반의 멀티캐스트 메커니즘들은 트리의 빈번한 재구성 및 자원 사용의 비효율적인 문제점을 가지고 있다. 이를 해결하기 위하여 제안된 오버레이 멀티캐스트 메커니즘들 역시 오버레이 트리의 재구성으로 인한 과부하와 밀접한 멤버간의 비효율적인 전송, 그리고 네트워크 확장성에 대한 문제점들을 갖는다. 본 논문에서는 이를 개선하기 위해 위치정보를 기반으로 밀접한 멤버노드들을 클러스터로 관리하여 클러스터 내에서는 브로드캐스팅을 적용하고, 멀티캐스트 소스와 클러스터 대표노드들간에는 유니캐스팅을 적용하는 오버레이 멀티캐스트 메커니즘을 제안한다.