• 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.

• The KIPS Transactions:PartA
• /
• v.9A no.2
• /
• pp.147-162
• /
• 2002

Multi-processor system with fast I/O devices improves processing performance and reduces the bottleneck by I/O concentration. In the system, the Performance influenced by shared memory used for exchanging data between processors varies with configuration and utilization. This paper suggests a prediction model for buffer and shared memory optimization under interrupt recognition method using mailbox. Ethernet (IEEE 802.3) packets are used as the input of system and the amount of utilized memory is measured for different network bandwidth and burstiness. Some empirical studies show that the amount of buffer and shared memory varies with packet concentration rate as well as I/O bandwidth. And the studies also show the correlation between two memories.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.2
• /
• pp.499-514
• /
• 1997

This paper proprses the architeecture of an interconnection network for Advanced Information Communi-cation Procssing System(AICPS)developde for prividing open information communication servies on a variety of heterogeneous networks.The proposed Interconnection network,called High Speed Swiching Fabric(HSSF),has been designed by a common bus.It can handile 32 i/O channels,each of which uses serial communication method using 100Mbps TAXI.The switching bandwidth of the common bus is 640Mvps.Each I/O channel can be alloted about 20Mbps bandwidth in steady state,and therefore it's sufficient bandwidth is able to interwork with ISDN and Internet services, as well as PSTN. HSSF is composed of the switching board assembly,the subscriber,I/O board assemly,and the backplane board assembly.An attached node takes in the network adapter board assembly to adapt the high speed interworking protocol.For reliability,HSSF is duplicated with load-sharing method.

• Korean Journal of Optics and Photonics
• /
• v.5 no.1
• /
• pp.74-79
• /
• 1994

The characteristics of the type I and type II SHG in LiB305 crystals grown by TSSG method have been investigated using 1064 nm beam from a Q-switched Nd:YAG laser. The measured phase matching angles and angular acceptance bandwidths were $\theta_m=90^{\circ}, \phi_m=11.6^{\circ}$, <$\delta\theta_{int}L_{1/2}=3.3^{\circ}-cm^{1/2}, \theta\phi_{int}L=0.27^{\circ}-cm^{1/2}$ for type I SHG and $\theta_m=20^{\circ}, \phi_m=90^{\circ}$, TEX>$\delta\theta_{int}L_=0.65^{\circ}-cm, \theta\phi_{int}L^{1/2}=3.5^{\circ}-cm^{1/2}$ for type II SHG, respectively. Thp. type I NCPM temperature of 1064 nm beam was found to be $149^{\circ}C$ with the temperature bandwidth $\DeltaTL$of $4.8^{\circ}C-cm$. An energy conversion efficiency of about 1.8% with 2.6 mm thick LBO crystal at an incident power of TEX>$171 MW/\textrm{cm}^2$ was demonstrated. The measured $d_{32} was 0.74\pm0.05 pm/V$..

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.676-679
• /
• 2015

빅데이터 분석을 위한 Hadoop 환경에서 고성능 저장장치인 SSD에 대한 중요성이 증가하면서 일반적으로 사용되는 저장장치인 HDD와 혼합하여 사용하는 연구들이 주목 받고 있다. 특히 SSD를 HDD의 캐시로 사용했을 때 저장장치에 대한 I/O 성능을 향상할 수 있다는 연구 결과들이 있다. 본 연구는 이를 바탕으로 SSD를 HDD의 캐시로 사용한다. HDFS는 저장장치에 접근하여 I/O를 수행하는데 기존에는 로컬 서버에서 캐시 미스가 발생한 경우 로컬 HDD로 접근한다. 이러한 방식은 접근하는 데이터에 따라 SSD의 높은 Bandwidth를 활용하지 못하게 되는 경우를 발생시키고 그 결과 특정 서버의 I/O 지연으로 전체 분산 처리의 성능을 저하시킬 수 있다. 이를 해결하기 위해 본 연구는 HDFS 레벨에서 로컬 서버의 HDD와 데이터 복제본들이 저장된 원격 서버의 SSD에서 I/O를 수행하는 경우에 대해 수식을 통해 비용을 비교한다. 그 결과 항상 기대 성능이 높은 저장 장치를 선택하여 데이터를 읽어오게 함으로써 기존 방식보다 성능이 개선될 수 있음을 입증한다.

• Journal of Industrial Technology
• /
• v.24 no.B
• /
• pp.177-186
• /
• 2004

The main bottleneck for a VOD system is bandwidth of storage or network I/O due to the high bandwidth requirements and long-lived nature of digital video. Patching is one of the most efficient techniques to overcome the bottleneck of the VOD system through the use of multicast scheme. In this paper, we propose a new patching scheme, P2P proxy patching, for improving the typical patching technique by jointly using the proxy prefix caching scheme and the P2P proxy. In our proposed scheme, each client play a role in a proxy for multicasting a regular stream to other clients that request the same stream. Due the use of the P2P proxy and the prefix caching, the server bandwidth is required significantly less than that of the typical patching technique. In the performance study, we show that our patching scheme can reduce the server bandwidth requirements compared with the existing patching techniques.

• ETRI Journal
• /
• v.37 no.5
• /
• pp.906-916
• /
• 2015

High-speed I/O channels require adaptive techniques to optimize the settings for filter tap weights at decision feedback equalization (DFE) read channels to compensate for channel inter-symbol interference (ISI) and crosstalk from multiple adjacent channels. Both ISI and crosstalk tend to vary with channel length, process, and temperature variations. Individually optimizing parameters such as those just mentioned leads to suboptimal solutions. We propose a joint optimization technique for crosstalk cancellation (XTC) at DFE to compensate for both ISI and XTC in high-speed I/O channels. The technique is used to compensate for between 15.7 dB and 19.7 dB of channel loss combined with a variety of crosstalk strengths from $60mV_{p-p}$ to $180mV_{p-p}$ adaptively, where the transmit non-return-to-zero signal amplitude is a constant $500mV_{p-p}$.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.736-743
• /
• 2016

Several studies have reported the impact of core affinity on the network I/O performance of multi-core systems. As the network bandwidth increases significantly, it becomes more important to determine the effective core affinity. Although a framework for dynamic core affinity that considers both network and disk I/O has been suggested, the multiple queues provided by high-speed I/O devices are not properly supported. In this paper, we extend the existing framework of dynamic core affinity to efficiently support the multiple queues of high-speed I/O devices, such as 40 Gigabit Ethernet and NVM Express. Our experimental results show that the extended framework can improve the HDFS file upload throughput by up to 32%, and can provide improved scalability in terms of the number of cores. In addition, we analyze the impact of the assignment policy of multiple I/O queues across a number of cores.

• Proceedings of the IEEK Conference
• /
• 2003.07b
• /
• pp.811-814
• /
• 2003

Today's and tomorrow's processors and I/O devices are demanding much higher I/O bandwidth than PCI 2.3 or PCI-X can deliver and it is time to engineer a new generation of PCI to serve as a standard I/O bus for future generation platforms. According to this demand the PCI SIG proposed PCI Express. This paper describes about the design of PCI Express Behavioral Model. A Behavioral Model enables the designers to test whether the design specifications are met by performing computer simulations rather than experiments on the physical prototype. In the proposed Model, we can verify whether our design concept satisfies the PCI Express functional protocol.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.685-688
• /
• 2001

고속 I/O 를 갖는 Loosely coupled 다중 프로세서 시스템은 데이터의 처리 성능 향상과 I/O 집중화에 따른 병목 현상을 줄여줄 수 있다. 이 때 프로세서간의 데이터 전송에 사용되는 공유메모리는 그 구성과 이용 방법에 따라 시스템의 성능에 많은 영향을 미친다. 본 연구에서는 공유메모리의 사용 방법을 비동기, 메일박스를 통한 인터럽트 전달 인지 방식으로, I/O 사용방법을 고속 이더넷(IEEE 802.3u)으로 한 시스템 모델에서 다중 프로세서 시스템 구성에 필요한 공유메모리의 최적 사용량을 입출력 데이터의 Bandwidth와 Burstness관점에서 실험을 통해 분석하였다.