• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.5_6
• /
• pp.290-298
• /
• 2003

Internet-based computer systems have to provide both high-availability and high-performance. Cluster technology has been used to obtain availability and performance simultaneously Generally, high-availability cluster systems tolerate a failure of a cluster node and cost-effectively solve it. In this paper, we study availability and downtime cost of (n,k) cluster systems. By considering performance, we model state transition of (n,k) cluster systems and apply software rejuvenation technique to improve availability of the system. We find that software rejuvenation can be used to improve availability of (n,k) cluster systems.

• Journal of Korean Institute of Industrial Engineers
• /
• v.11 no.2
• /
• pp.101-111
• /
• 1985

This study deals with problems of estimating the availability of the multi-module software systems. The result presented in this paper is an extension of our previous paper (2) entitled "A modified Markov model for the estimation of computer software performance". The extension is made by assuming that (1) the software system consists of R statistically independent software modules; (2) no failure occurrence while the process is in transition between software modules.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.2
• /
• pp.201-206
• /
• 2019

Rejuvenating a system periodically during the most idle time of the system reduces unexpected downtime caused by software aging and increases its availability. In general, software rejuvenation can be largely divided into two broad categories: time-based rejuvenation policy and condition-based rejuvenation policy. In time-based rejuvenation policy the software rejuvenation is triggered at scheduled time epochs with fixed time intervals, while in condition-based rejuvenation policy the software rejuvenation is performed when system state is checked to satisfy a specific condition. Conditionbased policy adds extra cost to the system due to system monitoring and aging estimation. This paper presents a stochastic model for analyzing time-based software rejuvenation mechanism, where the rejuvenation is triggered at scheduled time epochs with fixed time intervals, and provides an analytical solution for the steady-state availability, the user-perceived availability, and the corresponding cost.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.2
• /
• pp.227-234
• /
• 2002

An cluster web server system, one has the problem that does the low availability occured by the high chance of the server failures and it is not easy to provide high performance and availability occuring software aging. In this paper, running cluster web sewers consists of n primary servers and k backup servers, propose software rejuvenation model on performance and switchover time. Based on the various system operational parameters, we calculate to evaluate the rejuvenation policy such steady-state probabilities, availability, and downtime cost. And we validate the solutions of mathematical model by experiments based on various operation parameters and fud that the software rejuvenation method can be adopted as prventive fault tolerant technique for stability of system.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.7
• /
• pp.49-56
• /
• 2015

In this paper, we introduce recent cloud system security technologies categorizing them according to Reliability, Availability, Serviceability, Integrity, and Security (RASIS), terms that evaluate robustness of the computer system. Then we describe examples of security attacks and corresponding security technologies for each of them. We introduce security technologies based on Software Defined Network (SDN) for Reliability, security technologies based on hypervisor and virtualization for Availability, disaster restoration systems for Serviceability, authorization and access control technologies for Integrity, and encryption algorithms for Security. We believe that this paper provide wise view and necessary information for recent cloud system security technologies.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.2
• /
• pp.87-93
• /
• 2009

In this paper, we propose and evaluate the availability of single sensor node using a hierarchial modeling approach. We divides a sensor node into a software and hardware and analyze failures of each component. We construct Markov chains to represent the components of a sensor node, and then we construct a hierarchical model which use fault tree in upper level and Markov chains in lower level. We evaluate the availability and down of single sensor node.

• Communications for Statistical Applications and Methods
• /
• v.18 no.6
• /
• pp.741-750
• /
• 2011

In this paper, we derive an software mixed debugging model based on a Markov process, assuming that the length of time to perform the debugging is random and its distribution may depend on the fault type causing the failure. We assume that the debugging process starts as soon as a software failure occurs, and either a perfect debugging or an imperfect debugging is performed upon each fault type. One type is caused by a fault that is easily corrected and in this case, the perfect debugging process is performed. An Imperfect debugging process is performed to fix the failure caused by a fault that is difficult to correct. Distribution of the first passage time and working probability of the software system are obtained; in addition, an availability function of a software system which is the probability that the software is in working at a given time, is derived. Numerical examples are provided for illustrative purposes.

• Proceedings of the Korea Contents Association Conference
• /
• 2008.05a
• /
• pp.547-550
• /
• 2008

Recently, changes in the environment of user-centric internet service such as blog, UCC and IPTV and ubiquitous computing based on web service are needed to high availability system platform. High availability system is to provide safe service continuously even if system failure occurs in clustering system at the network. And it is necessary to synchronize data for reliable service in high availability system. In this paper, I implement DRBD(Disk Replicated Block Device) which is synchronization technique for data of local disks in high availability system.

• The KIPS Transactions:PartA
• /
• v.8A no.2
• /
• pp.81-90
• /
• 2001

As the rapid growth of Internet, computer systems are growing in its size and complexity. To meet high availability requirements for the systems, one usually uses both hardware and software fault tolerance techniques. To prevent failures of computer systems from software-aging phenomenon that come from long mission time, we adopt software rejuvenation method that stops and restarts the software in the servers intentionally. The method makes the systems clean and healthy state in which the probability of fault occurrence is very low. In this paper, we study how switchover time affects software rejuvenation of primary-backup server systems. Through experiments, we find that switchover time is an essential factor for deciding the rejuvenation policy.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.88-88
• /
• 2000

As more and more critical commercial applications move on the Internet, providing highly available servers becomes increasingly important. One of the advantages of a clustered system is that it has hardware and software redundancy. High availability can be provided by detecting node or daemon failure and reconfiguring the system appropriately so that the workload can be taken over bi the remaining nodes in the cluster. This paper presents how to provide the guaranteeing high availability of clustering web server. The load balancer becomes a single failure point of the whole system. In order to prevent the failure of the load balancer, we setup a backup server using heartbeat, fake, mon, and checkpointing fault-tolerance method. For high availability of file servers in the cluster, we setup coda file system. Coda is a advanced network fault-tolerance distributed file system.