• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06a
• /
• pp.203-205
• /
• 2012

최근 컴퓨터 시스템에서는 동작 주파수 증가에 따른 전력 소모량과 높은 온도문제로 인해 CPU의 성능에만 의존할 수는 없는 상황이다. 이에 따라 GPU 병렬처리 연산능력을 CPU의 범용 데이터 처리에 이용하는 기술에 대한 관심이 높아지고 있다. 하지만 CPU와 GPU의 모든 자원을 활용하기에는 이에 따른 높은 온도와 전력 상승이 문제가 된다. 따라서 본 논문에서는 GPU의 전력효율과 성능 측면에서 최적이 되는 동작 주파수에 대한 분석을 수행하고자 한다. GPU를 활용하는 API인 CUDA를 이용하여 GPU의 동작 주파수 변화에 따른 성능 변화, 전력 변화 그리고 Energy Delay에 대해서 분석한다. 실험을 통한 분석 결과 동작 주파수의 증가에 따라 성능은 최대 30%이상 증가했고, 전력소모량은 최대 약18%의 증가를 보여주었다. 또한 Energy Delay도 최대 21% 향상되는 것을 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.9
• /
• pp.2213-2221
• /
• 2015

The ondemand governor used in android-based smartphone platforms is a CPU frequency scaling technique. The ondemand governor sets the CPU operating frequency depending on the CPU utilization rate. Job scheduling affects the CPU utilization rate. The power consumption is proportional to the value of operating frequency. Consequently, CPU frequency scaling and CPU utilization rate have an effect on power consumption in a smartphone. In this paper, we evaluated the performance of job scheduling techniques incorporating the ondemand governor in terms of CPU utilization, power consumption, and job deadline miss ratio.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.526-529
• /
• 2015

as the smartphone industry developed, the smartphone's internal hardware devices have become high-end devices and it requires more power consumption than the previous one. therefore a battery of high capacity needed, but there is a limit in order to equip a large battery on account of smartphone minimization. The Linux Kernel provides the DVFS Mechanism to compensate for these limitations by software techniques. DVFS is dynamically adjust the frequency of the CPU to reduce the power consumption of the CPU. ondemand governor, the default policy in DVFS, apply the maximum frequency of the CPU whenever exceeding the up_threshold. so it result in a waste of CPU resources. by paying attention to this point, this paper propose the mechanism that maintain a high CPU utilization in proportion to the current frequency of the cpu to prevent the waste of CPU resources and conserve energy.

• Journal of Korea Game Society
• /
• v.17 no.4
• /
• pp.137-148
• /
• 2017

Since the power consumption of data centers is large and computer serves take a large portion of it, there have been much research on the power saving of servers in various ways recently. Among the units of severs CPU is one of major power consuming units. In this paper, a method of client-server assignment for minimizing the CPU power consumption of servers in a game server cluster is proposed. We model the client-server assignment problem as an optimization problem, and find a solution to the problem using a simulated annealing-based technique. One of major features of our method is to select a proper operating frequency according to the amount of load on a server. The selection of a lower frequency in case of low load will result in reducing power consumption. To our survey, little research on client-server assignment in consideration of power consumption has been carried out.

• Journal of Korea Society of Industrial Information Systems
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2011

Efficient power management plays a crucial role to strengthen competitiveness in the market of portable mobile commodities. This paper presents a proactive power management technique, called by Energy-Aware Scheduling policY (EASY), to exploit the sleep time information of running applications. Different from previous power management approaches focusing on power conservation in standby mode, the proposed scheme characterizes each application program's operational characteristic in active mode by observing how long the task stays in sleep state of CPU scheduler. Based on the measured sleep time, the proposed EASY speculates an adequate CPU clock frequency according to the current CPU workload and scales the frequency directly to the predicted one. Experimental results show that the proposed scheme reduces the power consumption by 10-30% on average compared to traditional DPM approach, with a minimal impact on the performance overhead.

• The Journal of Korean Association of Computer Education
• /
• v.16 no.3
• /
• pp.99-105
• /
• 2013

The FPGA-based emulation is an essential step in ASIC design for validation. For emulation with maximal frequency, it is crucial to understand the FPGA characteristics. This paper attempts to analyze the performance characteristics of the modern FPGAs from renowned vendors, Xilinx and Altera, with a case study utilizing various adders and MIPS CPU. Unlike the common wisdom, ripple-carry adder (RCA) does not utilize the inherent carry-chain inside FPGAs when structurally designed based on 1-bit adders. Thus, the RCA shows the inferior performance to the other types of adders in FPGAs. Our study also reveals that FPGAs from Xilinx exhibit different characteristics from the ones from Altera. That is, the prefix adder, which is optimized for speed in ASIC design, shows the poor performance on Xilinx devices, whereas it provides a comparable speed to the IP core on Altera devices. It suggests that error-prone manual change of the original design can be avoided on Altera devices if area is permitted. Experiments with MIPS CPU confirm the arguments.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06d
• /
• pp.4-6
• /
• 2012

모바일 환경에서는 배터리 용량의 한계로 에너지 효율성이 매우 중요하나, CPU의 이용률에 기반한 DVFS로는 소비 전력 감소에 한계가 존재한다. 모바일 어플리케이션은 사용자와의 상호 동작에 따라 다른 성향을 보여주기 때문에, 이 특성을 이용하여 사용자가 체감하는 성능저하 없이 소비 전력을 줄일 수 있다. 본 논문에서는 사용자 입력과 어플리케이션 간의 상관 관계를 바탕으로 수행 단계를 분석하였으며, 이에 맞추어 CPU 동작 주파수 범위를 조절함으로써 사용자 반응성을 유지하면서 소비 전력을 줄이는 방법을 제안한다. 이를 안드로이드 플랫폼에 적용하여 실험한 결과, 기존 대비 휴대폰의 이용 시간을 최대 2시간 증가시켰다.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1172-1175
• /
• 2020

Automotive ECU integrates CPU core, IVN controller, memory interface, sensor interface, I/O interface, and so on. Current automotive ECUs are often developed with proprietary processor architectures. However, demends for standard processors such as ARM and RISC-V increase rapidly for saftware compatibility in autonomous vehicles and connected cars. In this paper, an automotive ECU is designed for parking assist system based on RISC-V with open instruction set architecture. It includes 32b RISC-V CPU core, IVN controllers such as CAN and LIN, memory interfaces such as ROM and SRAM, and I/O interfaces such as SPI, UART, and I2C. Fabricated in 65nm CMOS technology, its operating frequency, area, and gate count are 50MHz, 0.37㎟, and 55,310 gates, respectively.

• 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.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.5
• /
• pp.9-19
• /
• 2012

As the clock frequency increases, CPU performance improves continuously. However, power and thermal problems in the CPU become more serious as the clock frequency increases. For this reason, utilizing the GPU to reduce the workload of the CPU becomes one of the most popular methods in recent high-performance computer systems. The GPU is a specialized processor originally designed for graphics processing. Recently, the technologies such as CUDA which utilize the GPU resources more easily become popular, leading to the improved performance of the computer system by utilizing the CPU and GPU simultaneously in executing various kinds of applications. In this work, we analyze the temperature and the energy efficiency of the computer system where the CPU and the GPU are utilized simultaneously, to figure out the possible problems in upcoming high-performance computer systems. According to our experimentation results, the temperature of both CPU and GPU increase when the application is executed on the GPU. When the application is executed on the CPU, CPU temperature increases whereas GPU temperature remains unchanged. The computer system shows better energy efficiency by utilizing the GPU compared to the CPU, because the throughput of the GPU is much higher than that of the CPU. However, the temperature of the system tends to be increased more easily when the application is executed on the GPU, because the GPU consumes more power than the CPU.