• Smart Media Journal
• /
• v.10 no.1
• /
• pp.25-31
• /
• 2021

As the machine learning becomes more common, development of application using machine learning is actively increasing. In addition, research on machine learning platform to support development of application is also increasing. However, despite the increasing of research on machine learning platform, research on suitable load balancing for machine learning platform is insufficient. Therefore, in this paper, we propose a load balancing scheme that can be applied to machine learning distributed environment. The proposed scheme composes distributed servers in a level hash table structure and assigns machine learning task to the server in consideration of the performance of each server. We implemented distributed servers and experimented, and compared the performance with the existing hashing scheme. Compared with the existing hashing scheme, the proposed scheme showed an average 26% speed improvement, and more than 38% reduced the number of waiting tasks to assign to the server.

• Journal of Communications and Networks
• /
• v.11 no.4
• /
• pp.375-383
• /
• 2009

The WiMedia alliance has specified a distributed medium access control (WiMedia MAC) protocol based on ultra wideband (UWB) for high data rate WPANs (HR-WPANs). The merits of WiMedia MAC such as distributed nature and high data rate make it a favorite candidate in HR-WPAN. Although QoS parameters such as the range of service rates are provided to a traffic stream, the WiMedia MAC is not able to use the QoS parameters and to determine or adjust a service rate using the QoS parameters for the traffic stream. In this paper, we propose a fair and adaptive resource allocation method that allocates time slots to isochronous streams according to QoS parameters and the current traffic load condition in a fully distributed manner. Although the traffic load condition changes, each device independently recognizes the changes and calculates fair and maximum allowable service rates for traffic streams. From the numerical and simulation results, it is proved that the proposed method achieves high capacity of traffic streams and fair QoS provisioning under various traffic load condition.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.6
• /
• pp.17-24
• /
• 2017

In recent digital broadcasting, up to 250 channels are multiplexed and transmitted. The channel transmission is made in the form of MPEG-2 Transport Stream (TS) and transmits one channel (Video, Audio). In order to check if many broadcast channels are transmitted normally, in multi-channel multi-view system, ability of real-time monitoring is required. In order to monitor efficient multi-channel, a distributed system in which functions and load are distributed should be implemented. In the past, we used an inefficient system that gave all of the functionality to a piece of hardware, which limited the channel acceptance and required a lot of resources. In this paper, we implemented a distributed multi-view system which can reduce resources and monitor them economically through efficient function and load balancing. It is able to implement efficient system by taking charge of decoding, resizing and encoding function in specific server and viewer function in separate server. Through this system, the system was stabilized, the investment cost was reduced by 19.7%, and the wall monitor area was reduced by 52.6%. Experimental results show that efficient real-time channel monitoring for multi-channel digital broadcasting is possible.

• Steel and Composite Structures
• /
• v.21 no.5
• /
• pp.1105-1119
• /
• 2016

In this study, it is presented that a new developed approach for equivalent area-distributed loading (EADL) induced from a single concentrated force. For the purpose, a full scale 3D steel formwork system was constructed in laboratory conditions. A developed load transmission platform was put on the formwork system and loaded step by step on the mass center. After each load increment, displacement was measured in several crictical points of the system. The developed platform which was put in to slab of formwork to equivalently distribute the load from a point to the whole slab was constituted using I profiles. A 3D finite element model of the formwork system was analyzed to compare numerical displacement results with experimental ones. In experimental tests,difference among the displacements obtained from reference numerical model (model applied EADL) and main numerical model (model applied single load using a load cell via load transmission platform) is about %13 in avarage. Difference among the displacements obtained from experimental results and main numerical model under 30 kN single load is about %11 in avarage. The results revealed that the displacements obtained experimentally and numerically are dramatically closed to each other. It is highlighted from the study that the developed approach is reliable and useful to get EDL.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.4
• /
• pp.1386-1396
• /
• 2017

In this paper, a new approach is proposed to select the optimal sitting and sizing of distributed solar photovoltaic generation (SPVG) in a radial electrical distribution systems (EDS) considering load/generation uncertainties. Here, distributed generations (DGs) allocation problem is modeled as optimization problem with network loss based objective function under various equality and inequality constrains in an uncertain environment. A boundary power flow is utilized to address the uncertainties in load/generation forecasts. This approach facilitates the consideration of random uncertainties in forecast having no statistical history. Uncertain solar irradiance is modeled by beta distribution function (BDF). The resulted optimization problem is solved by a new Dynamic Harmony Search Algorithm (DHSA). Dynamic band width (DBW) based DHSA is proposed to enhance the search space and dynamically adjust the exploitation near the optimal solution. Proposed approach is demonstrated for two standard IEEE radial distribution systems under different scenarios.

• Journal of Information Processing Systems
• /
• v.10 no.2
• /
• pp.193-214
• /
• 2014

Load balancing is the major benefit of any distributed system. To facilitate this advantage, task duplication and migration methodologies are employed. As this paper deals with dependent tasks (DAG), we used duplication. Task duplication reduces the overall schedule length of DAG along-with load balancing. This paper proposes a new task duplication algorithm at the time of tasks assignment on various processors. With the intention of conducting proposed algorithm performance computation; simulation has been done on the Netbeans IDE. The mesh topology of a distributed system is simulated at this juncture. For task duplication, overall schedule length of DAG is the main parameter that decides the performance of a proposed duplication algorithm. After obtaining the results we compared our performance with arbitrary task assignment, CAWF and HEFT-TD algorithms. Additionally, we also compared the complexity of the proposed algorithm with the Duplication Based Bottom Up scheduling (DBUS) and Heterogeneous Earliest Finish Time with Task Duplication (HEFT-TD).

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.10
• /
• pp.4642-4647
• /
• 2011

Solution for elastic foundation of plane strain state was derived by the application of variational method. Functions of the transverse distribution of the displacements for the analysis were chosen as linear functions. Loading conditions considered for the analysis were concentrated load and distributed load. Under the loading condition of the concentrated load, surface displacement was decreased drastically as the distance from the point of the loading increased. Under the loading condition of the distributed load, surface displacements were more uniformly distributed beneath the loading area when the ratio of the half of the loading width to the depth(B/H) of the compressible layer was greater. The surface displacement was more quickly converged from the edge of the loading area as the ratio(B/H) increased.

• Steel and Composite Structures
• /
• v.19 no.6
• /
• pp.1333-1354
• /
• 2015

A higher order analytical solution for static analysis of a truncated conical composite sandwich panel subjected to different loading conditions was presented in this paper which was based on a new improved higher order sandwich panel theory. Bending analysis of sandwich structures with flexible cores subjected to concentrated load, uniform distributed load on a patch, harmonic and uniform distributed loads on the top and/or bottom face sheet of the sandwich structure was also investigated. For the first time, bending analysis of truncated conical composite sandwich panels with flexible cores was performed. The governing equations were derived by principle of minimum potential energy. The first order shear deformation theory was used for the composite face sheets and for the core while assuming a polynomial description of the displacement fields. Also, the in-plane hoop stresses of the core were considered. In order to assure accuracy of the present formulations, convergence of the results was examined. Effects of types of boundary conditions, types of applied loads, conical angles and fiber angles on bending analysis of truncated conical composite sandwich panels were studied. As, there is no research on higher order bending analysis of conical sandwich panels with flexible cores, the results were validated by ABAQUS FE code. The present approach can be linked with the standard optimization programs and it can be used in the iteration process of the structural optimization. The proposed approach facilitates investigation of the effect of physical and geometrical parameters on the bending response of sandwich composite structures.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.60 no.1
• /
• pp.111-120
• /
• 2018

The purpose of this study was to develop a ridge regression (RR) model to estimate BOD and TP load using runoff weighted value. The concept of runoff weighted value, based on distributed curve-number (CN), was introduced to reflect the impact of land covers on runoff. The estimated runoff depths by distributed CN were closer to the observed values than those by area weighted mean CN. The RR is a technique used when the data suffers from multicollinearity. The RR model was developed for five flow duration intervals with the independent variables of daily runoff discharge of seven land covers and dependent variables of daily pollutant load. The RR model was applied to Heuk river watershed, a subwatershed of the Han river watershed. The variance inflation factors of the RR model decreased to the value less than 10. The RR model showed a good performance with Nash-Sutcliffe efficiency (NSE) of 0.73 and 0.87, and Pearson correlation coefficient of 0.88 and 0.93 for BOD and TP, respectively. The results suggest that the methods used in the study can be applied to estimate pollutant load of different land cover watersheds using limited data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.2
• /
• pp.386-393
• /
• 2004

In order to maintain load balancing in distributed systems in a decentralized manner, every node should obtain workload information from all the nodes on the network. It requires $Ο({v^2})$ traffic overheads, where v is the number of nodes. This paper presents a new synchronous dynamic distributed load balancing algorithm for a ( v,k+1,1)-configured network topology, which is a kind of 2k regular graph, based on symmetric balanced incomplete block design, where v equals ${k^2}+k+1$. Our algorithm needs only Ο(v√v) message overheads and each node receives workload information from all the nodes without redundancy. And load balancing in this algorithm is maintained so that every link has same amount of traffic by √v for transferring workload information.