• Journal of Internet Computing and Services
• /
• v.10 no.1
• /
• pp.101-110
• /
• 2009

Multi-agent systems are mostly used to integrate the intelligent and distributed approaches to various systems for effective sharing of resources and dynamic system reconfigurations. Object replication is usually used to implement fault tolerance and solve the problem of unexpected failures to the system. This paper presents the intelligent multi-agent distributed platform based on the dynamic object group management and proposes an object search technique based on the proposed filtered k-means (Fk-means). We propose Fk-means for the search mechanism to find alternative objects in the event of object failures and transparently reconnect client to the object. The filtering range of Fk-means value is set only to include relevant objects within the group to perform the search method efficiently. The simulation result shows that the proposed mechanism provides fast and accurate search for the distributed object groups.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.12
• /
• pp.610-617
• /
• 2007

Incremental checkpointing, which is intended to minimize checkpointing overhead, saves only the modified pages of a process. However, the cumulative site of incremental checkpoints increases at a steady rate over time because a number of updated values may be saved for the same page. In this paper, we present a comprehensive overview of Pickpt, a page-level incremental checkpointing facility. Pickpt provides space-efficient techniques aiming to minimizing the use of disk space. For our experiments, the results showed that the use of disk space using Pickpt was significantly reduced, compared with existing incremental checkpointing.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.3
• /
• pp.141-152
• /
• 2007

In this paper, we present a network-oriented lightweight real-time system, which is composed of an event-driven operating system called the Embedded Lightweight Operating System (ELOS) and a generic multi hop ad hoc routing protocol suite. In the ELOS, a conditional preemptive FCFS scheduling method with a guaranteed time slot is designed for efficient real-time processing. For more elaborate configurations, we reinforce fault tolerance by supplementing semi-auto configuration using wireless agent nodes. The developed hardware platform is also introduced, which is a scalable prototype constructed using off-the-shelf components. In addition, in order to evaluate the performance of the proposed system, we developed a ubiquitous network test-bed on which several experiments with respect to various environments are conducted. The results show that the ELOS is considerably favorable for tiny ubiquitous networked systems with real-time constraints.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.10
• /
• pp.5156-5161
• /
• 2013

In distributed spatial data stream processing, processed tuples of downstream nodes are replicated to the upstream node in order to increase the utilization of distributed nodes and to recover the whole system for the case of system failure. However, while the data input rate increases and multiple downstream nodes share the operation result of the upstream node, the data which stores to output queues as a backup can be lost since the deletion operation delay may be occurred by the delay of the tuple processing of upstream node. In this paper, the adaptive upstream backup scheme based on operation throughput in distributed spatial data stream system is proposed. This method can cut down the average load rate of nodes by efficient spatial operation migration as it processes spatial temporal data stream, and it can minimize the data loss by fluid change of backup mode. The experiments show the proposed approach can prevent data loss and can decrease, on average, 20% of CPU utilization by node monitoring.

• The KIPS Transactions:PartA
• /
• v.18A no.6
• /
• pp.225-232
• /
• 2011

Restricted Hypercube-Like (RHL) graphs are a graph class that widely includes useful interconnection networks such as crossed cube, Mobius cube, Mcube, twisted cube, locally twisted cube, multiply twisted cube, and generalized twisted cube. In this paper, we show that for an m-dimensional RHL graph G, $m{\geq}4$, with an arbitrary faulty edge set $F{\subset}E(G)$, ${\mid}F{\mid}{\leq}m-2$, graph $G{\setminus}F$ has a hamiltonian path between any distinct two nodes s and t if dist(s, V(F))${\neq}1$ or dist(t, V(F))${\neq}1$. Graph $G{\setminus}F$ is the graph G whose faulty edges are removed. Set V(F) is the end vertex set of the edges in F and dist(v, V(F)) is the minimum distance between vertex v and the vertices in V(F).

• Journal of Internet Computing and Services
• /
• v.15 no.6
• /
• pp.1-8
• /
• 2014

As the e-learning spreads widely and demands on the internet video service, transmitting video data for many users over the Internet becomes popular. To satisfy this needs, the traditional approach uses a tree structure that uses the video server as the root node. However, this approach has the danger of stopping the video service even when one of the nodes along the path has a some problem. In this paper, we propose a video-on-demand service that uses multiple paths. We add new paths for backup and speed up for transmitting the video data. We show by simulation experiments that our approach provides a high-quality of video service.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.37 no.1
• /
• pp.19-28
• /
• 2000

The parallel inverter is widely utilized because of its fault-tolerance capability, high-current output at constant voltages and system modularity. The conventional paralled inverter usually employes an active and reactive power control or a frequency and voltage droop control. However, these approaches have the disadvantages that the response time of parallel inverter control is slow against load and system parameter variation to calculate active, reactive power, frequency and voltage. This paper describes novel control scheme for equalization of output power between the parallel connected inverters. The proposed scheme has a fast power balance control response, a simplicity of implementation, and inherent peak current limiting capability since it employes a instantaneous current/voltage control with output voltage and current balance and output voltage regulation. A design procedure for the proposed parallel inverter controller is presented. Futhermore, the proposed constrol scheme is verified through the simulation in various cases such as the system parameter variation, the control parameter variation and the nonlinear load condition.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.6
• /
• pp.33-44
• /
• 2008

Multipath routing protocols with load balance, fault tolerance, aggregated bandwidth in Ad-hoc Networks provide improved throughput and reliable route as compared with singlepath routing protocols. However, multipath routing protocols have not been explored thoroughly in the domain of overhead in Ad-hoc Networks. In this paper, we analyze and compare on-demand singlepath and multipath routing with IEEE 802.11 DCF in terms of Routing overhead and MAC overhead. The results reveals that in comparison with singlepath routing protocol, multipath routing mechanism creates more overheads but provides better performance in congestion and capacity provided that the route length is within a certain upper bound which is derivable. The analytical results are further confirmed by simulation.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.9
• /
• pp.445-453
• /
• 2003

As cluster systems used for high performance computing consist of large number of running servers, one has to solve the low availability problems occurred by the high chance of the server failures. To handle the problems, it is necessary to have the precise definition of available performance of cluster systems that represents availability and performability of the systems simultaneously. Previous research results that mention availability issues lack for concerning system performance such as waiting time and response time in their availability definition. In this paper, we propose a new availability metric for (n, k)-way cluster systems which compose of n primary servers and k backup servers. With the metric, the change of system performance according to arrival rates is captured and the waiting time of a request can be kept below to a certain level. Using various system operating parameters, we calculate availability and downtime of cluster systems along with waiting tine deadline.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1303-1313
• /
• 2015

MapReduce provides high levels of system scalability and fault tolerance for large-size data processing. A MapReduce-based k-nearest-neighbor(k-NN) join algorithm seeks to produce the k nearest-neighbors of each point of a dataset from another dataset. The algorithm has been considered important in bigdata analysis. However, the existing k-NN join query-processing algorithm suffers from a high index-construction cost that makes it unsuitable for the processing of bigdata. To solve the corresponding problems, we propose a new grid-based, k-NN join query-processing algorithm. Our algorithm retrieves only the neighboring data from a query cell and sends them to each MapReduce task, making it possible to improve the overhead data transmission and computation. Our performance analysis shows that our algorithm outperforms the existing scheme by up to seven-fold in terms of the query-processing time, while also achieving high extent of query-result accuracy.