• Journal of Digital Contents Society
• /
• v.18 no.4
• /
• pp.747-751
• /
• 2017

The log has variable informations that are important and necessary to manage a network when accessed to network servers. These informations are used to reduce a cost and efficient manage a network through the meaningful prediction information extraction from the amount of user access. And, the network manager can instantly monitor the status of CPU, memory, disk usage ratio on network using the SNMP. In this paper, firstly, we have accumulated and analysed the 6 network logs and extracted the informations that used to predict the amount of user access. And then, we experimented the prediction simulation with the time series analysis such as moving average method and exponential smoothing. Secondly, we have simulated the usage ration of CPU, memory, and disk using Xian SNMP simulator and extracted the OID for the time series prediction of CPU, memory, and disk usage ration. And then, we presented the visual result of the variable experiments through the Excel and R programming language.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.6
• /
• pp.57-67
• /
• 2010

This paper studies thermal characteristics of a mix of CPU-intensive and memory-intensive application workloads on a multi-core processor. Especially, we focus on thermal variations during program execution because thermal variations are more critical than average temperatures and their ranges for thermal management. New metrics are proposed to quantify such thermal variations for a workload. We study the thermal variations using SPEC CPU2006 benchmarks with varying cooling conditions and frequencies on an Intel Core 2 Duo processor. The results show that applications have distinct thermal variations characteristics. Such variations are affected by cooling conditions,operating frequencies and multiprogramming workload. Also, there are distinct spatial thermal variations between cores. Our new metrics and their results from this study provide useful insight for future research on multi-core thermal management.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.9
• /
• pp.189-196
• /
• 2020

Recently, GPU cloud computing technology applying GPU(Graphics Processing Unit) devices to virtual machines is widely used in the cloud environment. In a cloud environment, GPU devices assigned to virtual machines can perform operations faster than CPUs through massively parallel processing, which can provide many benefits when operating high-performance computing services in a variety of fields in a cloud environment. In a cloud environment, a GPU device can help improve the performance of a virtual machine, but the virtual machine scheduler, which is based on the CPU usage time of a virtual machine, does not take into account GPU device usage time, affecting the performance of other virtual machines. In this paper, we test and analyze the performance degradation of other virtual machines due to the virtual machine that performs GPGPU(General-Purpose computing on Graphics Processing Units) task in the direct path based GPU virtualization environment, which is often used when assigning GPUs to virtual machines in cloud environments. Then to solve this problem, we propose a GPGPU task management method for a virtual machine.

• 한국정보통신설비학회:학술대회논문집
• /
• 2006.08a
• /
• pp.265-268
• /
• 2006

This paper, suggest the Rrecord Management solution for unified RFID leader System based on WindowsCE. The unified RFID Reader System consider of RFID Module, GPS Module, CPU Module, WLAN(or CDMA) Module and DB into Local DB and Remote DB for of effective Record Management Systems.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.159-164
• /
• 2008

The efficiently power management Is an important requirement traditionally in the mobile communication system which uses battery as their power source. Especially, it has been emphasized in the most recent devices, which has to provide high performance and various functions with an extended operating time. In this article, the adaptive Power management technique for the core processor unit in embedded systems used widely for the mobile system thanks to its advantage on power consumption and physical site, is proposed.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.27 no.2
• /
• pp.153-171
• /
• 2002

In this paper, we present an efficient pricing strategy, the pivot and probe Algorithm for Network Flow Problems(PAPANET), specifically for solving capacitated, linear network flow problem (NPs). The PAPANET begins with an initial relaxed network problem(RNP), consisting of all the nodes and initial candidate arcs(possibly a few least cost arcs form the original problem and a set of all the artificial and slack arcs). After an initial solution to the RNP is derived by pivoting, the PROBE procedure identifies a set of most violated arcs from the noncandidate arcs that are not considered to be in the current RNP, and adds them to the RNP. The procedure also discards a set of least favorable, zero flow, nonbasic arcs from the RNP. The new RNP is solved to optimality and the procedure continues until all of the dual constraints of the noncandidate arcs are satisfied by the dual solution to the RNP. The PAPANET effectively reduces the problem size, time per pivot, and solution CPU time by eliminating noncandidate arcs. Computational tests on randomly generated problems indicate that PAPANET achieves and average savings of 50-80% of the solution CPU time of that of a comparable standard network simplex implementation.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.8
• /
• pp.13-18
• /
• 2016

In Internet of Things (IoT) environment, a variety of smart devices are connected to Internet through various network technologies such as 4G/5G, WiBro, Bluetooth, etc. in order to provide the remote monitoring service such as smart healthcare service, etc. Most standard mobility management protocols based on IETF Mobile IP may not be suitable for Internet of Things (IoT) environments due to constrained power, constrained CPU processing and memory capacity, and large signalling overhead which are inherently accompanied by various devices in IoT environments. In this article, we propose a new mobility management protocol CoMP (CoAP-based Mobility Managemenbt Protocol) for reliable mobility management in IoT environments. The architecture and algorithm to achieve both reliability and simplicity for IoT mobility management are proposed. Finally, performance has been evaluated by both mathematical analysis and simulation.

• Journal of The Korean Association of Information Education
• /
• v.3 no.1
• /
• pp.86-93
• /
• 1999

In this paper, we have designed and implemented that PC management system has the function of monitoring and controlling system management MIB defined for PC management using SNMP in PC room. System management MIB consists of monitoring MIB for system and processor information and controlling MIB for backup and restoration. For this, we have added new defined MIB variables to standard MIB and displayed system information, processor information, backup and restoration information of managed PC. And the structure between management system and managed system is shown and process of system information, processor information, backup and restoration of managed PC is illustrated. On real environmental, by monitoring window version, CPU type, system memory and virtual memory and performing backup and restoration of managed PC, actual operation is tested. Management system gathers system information and processor information from managed system and performs backup and restoration for recovering fault of managed system. Hence PC managers can operate and manage many PCs on LAN, efficiently.

• Journal of KIISE
• /
• v.43 no.2
• /
• pp.256-268
• /
• 2016

Modern networks consisting of various heterogeneous equipment are often installed in a distributed manner. Thus the NETCONF standard was established to manage networks centrally and efficiently. In this paper, we present a method that integrates each NETCONF layer into a single system based on the results of previous studies. In the RPC Layer, an asynchronous communication channel and parallel processes are possible using multi-threading. In the Operation Layer, operational efficiency is increased by using a data group with dependencies between the equipment configuration data and by improving the data structure, enabling efficiently processing of XML queries even with multiple managers. The data modeling techniques and grouping methods in the Content Layer are presented in detail for interoperability between the Operation Layer and the Content Layer. Finally, the GUI program was implemented and its implementation is reported. We performed an experiment comparing the improved NETCONF with the standard NETCONF to measure factors, such as query processing ratio, query processing speed, and CPU utilization. The improved NETCONF demonstrated excellent query processing ratio and query processing speed, whereas the standard NETCONF had excellent CPU utilization.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1405-1411
• /
• 2022

Power management is still an important issue in embedded environments as hardware advances like high-performance processors. Power management methods such as DVFS control CPU frequencies in an adaptive manner for efficient power management in polling-based I/O programs such as network communication. This paper presents the problems of the existing power management method and proposes a new power management method. Through this, it is possible to reduce electric consumption by increasing the polling cycle in situations where the frequency of data reception is low, and on the contrary, in situations where data reception is frequent, it can operate at the maximum frequency without performance degradation. After implementing this as a code layer on the embedded board and observing it through Atmel's Power Debugger, the proposed method showed a performance improvement of up to 30% in energy consumption compared to the existing power management method.