• Smart Media Journal
• /
• v.5 no.4
• /
• pp.18-25
• /
• 2016

Virtualized hardware resources provides efficiency in use and easy of management. Based on the benefits, virtualization techniques are used to build large server clusters and cloud systems. The performance of a virtualized system is significantly affected by the virtual machine scheduler. However, the existing virtual machine scheduler have a problem in that the I/O response is reduced in accordance with the scheduling delay becomes longer. In this paper, we introduce the Loan/Redemption mechanism of a virtual machine scheduler in order to improve the responsiveness to I/O events. The proposed scheme gives additional credits for to virtual machines and classifies the task characteristics of each virtual machine by analyzing the credit consumption pattern. When an I/O event arrives, the scheduling priority of a virtual machine is temporally increased based on the analysis. The evaluation based on the implementation shows that the proposed scheme improves the I/O response 60% and bandwidth of virtual machines 62% compared to those of the existing virtual machine scheduler.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.1
• /
• pp.167-174
• /
• 2009

A Tile Map Service is a regular map service that has been enhanced to serve maps very quickly using a cache of static images. The map cache is a directory that contains image tiles of a map extent at specific scale levels. Returning a tile from the cache takes the server much less time than drawing the map image on demand. Use of a Tile Map Service can dramatically improve the time that clients take to display complex base-maps. Using Tile Map Services thus eliminate the need to trade quality for performance. This study provides a way to construct Tile Map Service System using Open Source GIS. We used GDAL(Geospatial Data Abstraction Library) which is one of the Open Source GIS Softwares to make Tile Map Image and OpenLayers to publish Web Page. Moreover, We conducted a performance test on Tile Map System and Dynamic Map System and evaluated the results of it. As a result, the proposed method makes it easier to construct high performance Tile Map Service using Open Source GIS without commercial products.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.4
• /
• pp.1058-1070
• /
• 1999

In this paper, we propose a control model which can control the burst input messages of the BSC(Base Station controller) in mobile communication systems more efficiently and reliably, by dividing the input messages characteristically and using multiprocessor system. Using M/M/c/K queueing model, we briefly analyze proposed model to get characteristic parameters which are required to performance improvement. On the base of the results, we compare our proposed model with the conventional one by using SLAM II with regard to the following factors : the call blocking rate of the input message, the distribution of average queue length, the utilization of process controller(server), and the distribution of average waiting time in queue. In addition, we modified our model which has overload control function for burst input messages, and analyzed its performance.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1221-1230
• /
• 1997

In this study, the performance of a parallel file system(N-PFS), which is inplemented using conventional disks as disk arrays on a Workstation Cluster, is analyzed by using analytical method and adtual values in experiments.N-PFS can be used as high-performance file sever in small-scale server systems and effciently pro-cess massive data I/Os such as multimedia and scientifid data. In this paper, an analytical model was suggested and the correctness of the suggested was verified by analyzing the experimental values on a system.The result of the appropriate stping unit for processing massive data of the Workstation Cluster with 8 disks is 64-128Kbytes and the maximum throughput on it is 15.8 Mbytes/ses.In addition, the performance of parallel file system on massive data is bounded by the time required to copy data between buffers.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.9
• /
• pp.17-28
• /
• 2004

Since the Internet has grown into a large communication infrastructure, the network management system to systematically manage the large number of network components has become an essential part of the Internet. The rapid growth of network size has brought into question the scalability of the existing centralized model, such as SNMP (Simple Network Management Protocol) and CMIP (Common Management Information Protocol). Thus, for efficient network management, researches about mobile agent have also been performed recently. This paper presents analytical models of centralized approach based on SNMP protocol, distributed approach based on mobile agent, and mixed mode to make up for shortcomings of SNMP and mobile agent. We compare the performance of these analytical models based on network management response time. Experiment results show that performance of mobile agent and the mixed mode is less sensitive to the delay in WAN network environment. However, SNMP is more efficient for the simple network environment like LAN.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.820-826
• /
• 2016

There is a growing requirement for big data processing which extracts valuable information from a large amount of data. The Hadoop system employs the MapReduce framework to process big data. However, MapReduce has limitations such as inflexible and slow data processing. To overcome these drawbacks, SQL query processing techniques known as SQL-on-Hadoop were developed. Apache Tajo, one of the SQL-on-Hadoop techniques, was developed by a Korean development group. External merge sort is one of the heavily used algorithms in Tajo for query processing. The performance of external merge sort in Tajo is influenced by two parameters, sort buffer size and fanout. In this paper, we analyzed the performance of external merge sort in Tajo with various sort buffer sizes and fanouts. In addition, we figured out that there are two major causes of differences in the performance of external merge sort: CPU cache misses which increase as the sort buffer size grows; and the number of merge passes determined by fanout.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.3
• /
• pp.617-630
• /
• 1996

In this paper, a shared-concurrent file system (S-CFS) is designed and implemented using conventional disks as disk arrays on a Workstation Cluster which can be used as a small-scale server. Since it is implemented on UNIX operating systems, S_CFS is not only portable and flexible but also efficient in resource usage because it does not require additional I/O nodes. The result of the research shows that on small-scale systems with enough disks, the performance of the concurrent file system on transaction processing applications is bounded by the bottleneck of CPUs computing powers while the performance of the concurrent file system on massive data I/Os is bounded by the time required to copy data between buffers. The concurrent file system,which has been implemented on a Workstation Cluster with 8 disks,shows a throughput of 388 tps in case of transaction processing applications and can provide the bandwidth of 15.8 Mbytes/sec in case of massive data processing applications. Moreover,the concurrent file system has been dsigned to enhance the throughput of applications requirring high performance I/O by controlling the paralleism of the concurrent file system on user's side.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.367-373
• /
• 2022

VANET (Vehicular Ad hoc NETwork) is one of the special cases of the ad hoc networks in which car nodes communicate with each other and/or with RSUs (Road Side Unit) in order for the drivers to receive nearby road traffic information as well as for the passengers to retrieve nearby gas price or hotel information. In case of constructing VANET over CCN, users do not need to specify a destination server address rather to input a key word such as nearby congestion in order to gather surrounding traffic congestion information. Furthermore, each car node caches its retrieved data for forwarding other nodes when requested. In addition, the data transmission is inherently multicast, which implies fast data propagation to the participating car nodes. This paper measures and evaluates the data transmission performance of the VCCN (VANET over CCN) in which nodes are equipped with diverse wireless communication channels. The simulation result indicates that 802.11a shows the best performance of the data transmission against other wireless channels. Moreover, it indicates that VCCN improves overall data transmission and provides benefit to the nodes that request the same traffic information by exploiting inherent multicast communication.

• KNOM Review
• /
• v.23 no.1
• /
• pp.43-50
• /
• 2020

As General Purpose Graphics Processing Unit (GPGPU) recently plays an essential role in high-performance computing, several cloud service providers offer GPU service. Most cluster orchestration platforms in a cloud environment using containers allocate the integer number of GPU to jobs and do not allow a node shared with other jobs. In this case, resource utilization of a GPU node might be low if a job does not intensively require either many cores or large size of memory in GPU. GPU virtualization brings opportunities to realize kernel concurrency and share resources. However, performance may vary depending on characteristics of applications running concurrently and interference among them due to resource contention on a node. This paper proposes GPU container co-execution framework with multiple server creation and execution based on Kubernetes, container orchestration platform for measuring interference which may be occurred by sharing GPU resources. Performance changes according to scheduling policies were investigated by executing several jobs on GPU. The result shows that optimal scheduling is not possible only considering GPU memory and computing resource usage. Interference caused by co-execution among applications is measured using the framework.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.2
• /
• pp.405-416
• /
• 2022

Current AI technology is improving the quality of life by using machine learning based on data. When using machine learning, transmitting distributed data and collecting it in one place goes through a de-identification process because there is a risk of privacy infringement. De-identification data causes information damage and omission, which degrades the performance of the machine learning process and complicates the preprocessing process. Accordingly, Google announced joint learning in 2016, a method of de-identifying data and learning without the process of collecting data into one server. This paper analyzed the effectiveness by comparing the difference between the learning performance of data that went through the de-identification process of K anonymity and differential privacy reproduction data using actual financial data. As a result of the experiment, the accuracy of original data learning was 79% for k=2, 76% for k=5, 52% for k=7, 50% for 𝜖=1, and 82% for 𝜖=0.1, and 86% for Federated learning.