• The KIPS Transactions:PartC
• /
• v.19C no.2
• /
• pp.135-144
• /
• 2012

All the servers in a traditional server cluster environment are kept On. If the request load reaches to the maximum, we exploit its maximum possible performance, otherwise, we exploit only some portion of maximum possible performance so that the efficiency of server power consumption becomes low. We can improve the efficiency of power consumption by controlling power mode of servers according to load situation, that is, by making On only minimum number of servers needed to handle current load while making Off the remaining servers. In the existing power mode control method, they used a static policy to decide server power mode at a fixed time interval so that it cannot adapt well to the dynamically changing load situation. In order to improve the existing method, we propose a dynamic server power control algorithm. In the proposed method, we keep the history of server power consumption and, based on it, predict whether power consumption increases in the near future. Based on this prediction, we dynamically change the time interval to decide server power mode. We performed experiments with a cluster of 30 PCs. Experimental results show that our proposed method keeps the same performance while reducing 29% of power consumption compared to the existing method. In addition, our proposed method allows to increase the average CPU utilization by 66%.

• KIPS Transactions on Computer and Communication Systems
• /
• v.4 no.11
• /
• pp.369-382
• /
• 2015

Energy aware server clusters aim to reduce power consumption at maximum while keeping QoS(quality of service) compared to energy non-aware server clusters. They adjust the power mode of each server in a fixed or variable time interval to activate only the minimum number of servers needed to handle current user requests. Previous studies on energy aware server cluster put efforts to reduce power consumption or heat dissipation, but they do not consider energy efficiency well. In this paper, we propose an energy efficient cluster management method to improve not only performance per watt but also QoS of the existing server power mode control method based on autonomous learning. Our proposed method is to adjust server power mode based on a hybrid approach of autonomous learning method with multi level thresholds and power consumption prediction method. Autonomous learning method with multi level thresholds is applied under normal load situation whereas power consumption prediction method is applied under abnormal load situation. The decision on whether current load is normal or abnormal depends on the ratio of the number of current user requests over the average number of user requests during recent past few minutes. Also, a dynamic shutdown method is additionally applied to shorten the time delay to make servers off. We performed experiments with a cluster of 16 servers using three different kinds of load patterns. The multi-threshold based learning method with prediction and dynamic shutdown shows the best result in terms of normalized QoS and performance per watt (valid responses). For banking load pattern, real load pattern, and virtual load pattern, the numbers of good response per watt in the proposed method increase by 1.66%, 2.9% and 3.84%, respectively, whereas QoS in the proposed method increase by 0.45%, 1.33% and 8.82%, respectively, compared to those in the existing autonomous learning method with single level threshold.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.3
• /
• pp.155-164
• /
• 2016

For frequency pattern analysis of large databases, the new tree-based frequency pattern analysis algorithm which can compensate for the disadvantages of the Apriori method has been variously studied. In frequency pattern tree, the number of nodes is associated with memory allocation, but also affects memory resource consumption and processing speed of the growth. Therefore, reducing the number of nodes in the tree is very important in the frequency pattern mining. However, the absolute criteria which need to order the transaction items for construction frequency pattern tree has lowered the compression ratio of the tree nodes. But most of the frequency based tree construction methods adapted the absolute criteria. FP-tree is typically frequency pattern tree structure which is an extended prefix-tree structure for storing compressed frequent crucial information about frequent patterns. For construction the tree, all the frequent items in different transactions are sorted according to the absolute criteria, frequency descending order. CanTree also need to absolute criteria, canonical order, to construct the tree. In this paper, we proposed a novel frequency pattern tree construction method that does not use the absolute criteria, IRFP-tree algorithm. IRFP-tree(Intersection Rule based FP-tree). IRFP-tree is constituted with the new paradigm of the intersection rule without the use of the absolute criteria. It increased the compression ratio of the tree nodes, and reduced the tree construction time. Our method has the additional advantage that it provides incremental mining. The reported test result demonstrate the applicability and effectiveness of the proposed approach.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.2
• /
• pp.443-453
• /
• 2013

The hybrid test is one of the most advanced test methods to predict the structural dynamic behavior with the interaction between a physical substructure and a numerical modeling in the hybrid control system. The purpose of this study is to perform the multi-directional dynamic test of a steel frame structure with the real-time hybrid system and to evaluate the validation of the results. In this study, FEAPH, nonlinear finite element analysis program for hybrid only, was developed and the hybrid control system was optimized. The inefficient computational time was improved with a fixed number iteration method and parallel computational techniques used in FEAPH. Furthermore, the previously used data communication method and the interface between a substructure and an analysis program were simplified in the control system. As the results, the total processing time in real-time hybrid test was shortened up to 10 times of actual measured seismic period. In order to verify the accuracy and validation of the hybrid system, the linear and nonlinear dynamic tests with a steel framed structure were carried out so that the trend of displacement responses was almost in accord with the numerical results. However, the maximum displacement responses had somewhat differences due to the analysis errors in material nonlinearities and the occurrence of permanent displacements. Therefore, if the proper material model and numerical algorithms are developed, the real-time hybrid system could be used to evaluate the structural dynamic behavior and would be an effective testing method as a substitute for a shaking table test.

• Journal of the Society of Naval Architects of Korea
• /
• v.36 no.4
• /
• pp.137-146
• /
• 1999

Inspection becomes to be important in the safety of structure and economical viewpoint, because structural damage accompanies lots of economical cost and social problems. Especially ship structure is composed of a lot of members and it is impossible to inspect all members continuously. The purpose of this paper is to get optimal inspection plan containing inspection time and method. Crack is one of major modes on the structural failure and can lead to collapse of structure. In this paper, the deteriorating process, which contains inspection to detect the crack before the propagation to large crack, is idealized as Markov chain model. Genetic algorithm is also used to accomplish the optimization of inspection plan. Especially, the probabilistic characteristics of cracks are changed, because ship is operating in corrosive environments and the scantling of structural members is reduced due to corrosion. Non-stationary Markov chain model is used to represent the process of corrosion in structural members. In this paper, the characteristics of indivisual inspection plan are compared by numerical examples for the change of corrosion rate, the cost due to scheduled system down and target failure probability. From the numerical example, it can be seen that the improvement of fatigue life for the members with short fatigue life is the most effective way in order to reduce total maintenance cost.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.2B
• /
• pp.98-106
• /
• 2006

In order to provide a realtime transmission over a wireless LAM, we here present a new jamming based retransmission mechanism. In a legacy wireless LAN system, all stations use the binary exponential backoff algorithm to avoid collisions among frames. It is well known that the backoff algorithm causes more collisions as the numbers of active stations increases. This makes transmission of real time traffic hard. In the proposed scheme, when each station senses collisions, it promptly allows to send a jamming signal during a unique jamming window period which is determined by its own channel access count database(CACDB). This jamming windows is chosen not to be overlapped each other by using of CACDB, and thus channel access of another station is prevented. Hereafter the station gets the ownership of the medium when the wireless medium becomes idle after sending the jamming signal and sensing carrier, and then sends frame in medium. In our proposal, repeating collisions is never happened. We here assume that real time traffic use a frame of fixed length in order to make the time for receiving its ACK frame same. Comparing the proposed jamming-based retransmission scheme with the the 802.11 and 802.11e MAC by simulation. one can find that the proposed scheme have advantages in terms of delay, average backoff time, and average number of collisions per frame. One can find that the proposed scheme might be practically applicable to several applications of realtime traffic transmission in wireless LAN systems.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.3
• /
• pp.197-205
• /
• 2020

Korean language has the characteristics that the pronunciation of phoneme units such as vowels and consonants are fixed and the pronunciation associated with a notation does not change, so that foreign learners can approach rather easily Korean language. However, when one pronounces words, phrases, or sentences, the pronunciation changes in a manner of a wide variation and complexity at the boundaries of syllables, and the association of notation and pronunciation does not hold any more. Consequently, it is very difficult for foreign learners to study Korean standard pronunciations. Despite these difficulties, it is believed that systematic analysis of pronunciation errors for Korean words is possible according to the advantageous observations that the relationship between Korean notations and pronunciations can be described as a set of firm rules without exceptions unlike other languages including English. In this paper, we propose a visualization framework which shows the differences between standard pronunciations and erratic ones as quantitative measures on the computer screen. Previous researches only show color representation and 3D graphics of speech properties, or an animated view of changing shapes of lips and mouth cavity. Moreover, the features used in the analysis are only point data such as the average of a speech range. In this study, we propose a method which can directly use the time-series data instead of using summary or distorted data. This was realized by using the deep learning-based technique which combines Self-organizing map, variational autoencoder model, and Markov model, and we achieved a superior performance enhancement compared to the method using the point-based data.

• Journal of Broadcast Engineering
• /
• v.17 no.1
• /
• pp.179-190
• /
• 2012

In Distributed Video Coding(DVC), a low complexity encoder can be realized by shifting complex processes of encoder such as motion estimation to decoder. Since not only motion estimation/compensation processes but also channel decoding process needs to be performed at DVC decoder, the complexity of a decoder is significantly increased in consequence. Therefore, various fast channel decoding methods are proposed for the most computationally complex part, which is the channel decoding process in DVC decoding. As the channel decoding process becomes faster using various methods, however, the complexity of the other processes are relatively highlighted. For instance, the complexity of side information generation process in the DVC decoder is relatively increased. In this paper, therefore, a fast method for the DVC decoding is proposed by using adaptive search range method in side information generation process. Experimental results show that the proposed method achieves time saving of about 63% in side information generation process, while its rate distortion performance is degraded only by about 0.17% in BDBR.

• Journal of KIISE:Information Networking
• /
• v.30 no.4
• /
• pp.545-558
• /
• 2003

It is expected that per-user customized services are widely used in next generation Personal Communication Network. The ultimate goal for personalized service is the Virtual home Environment (VHE) providing ´same-look-and-feel´ services for the subscriber wherever he roams to. To provide personalized services for each call, per-user service profiles are frequently referenced, so efficient service profile management is essentially required. To realized the VHE, typically two schemes, can be employed; One is Intelligent Network based service control and the other is a full replication scheme that always replicates profile in user´s current zone. The first scheme is referred as Central scheme and th second scheme is the modified replication scheme of IMT-2000, we refer to as Follow-Me Replication Unconditional (FMRU). Since the Central scheme only depends on the service cal rate and the FMRU is merely dependent on the movement rate, it is apparent that FMRU scheme outperforms the Central scheme if the call to mobility ratio (CMR) is large, and vice versa. In this paper, we propose a new service profile replication schemes, Adaptive Follow-Me Replication (AFMR) that determine replication automatically according to the user´s CMR. We compared the performance of the AFMR with the non-adaptive Follow-Me Replication unconditional on Demand (FMRUD) scheme. Performance results indicate that as the CMR of a user changes AFMR adapts well compared to the existing schemes.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.8
• /
• pp.2013-2026
• /
• 1998

In optical passive star interconnections, all packets are transmitted between nodes ina broadcast and-select manner. It is assumed that each node has a innable transmitter and a fixed-savelength receiver, ad that all packet lengths are equal so that each transmission can be done in a unit time. The tuning delay, denoted by $\delta$, means the amount of time for transmitter to change its wavelength to another one. The problec is , given ay value of the mumber of nodes N and the number of wavelengths $\kappa$ according to WDM implementations, to find transmission schedules with minimum cycle length for all-to all brondcaxt where no one sends any packet to itself. In this paper, we prove that the cycle length of optimal transcission schedules should be at least $max[[{\frac{N}{k}](N-1)}]$,$k\delta$$+N-1$. A novel algorithm for optimal transmission schedules is then presented when N-1 is divisible by $\kappa$. This algorithm also can be used for any values of N and $\kappa$ if the tuning delay $\delta$ does not affect strictly the cycle length of transmission schedules, i,e, $[\frac{N}{k}](N-1)$ > $\kappa\delta$+N-1.