• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.11
• /
• pp.1330-1336
• /
• 2019

As the Fourth Industrial Revolution era began, a variety of devices are running on the cloud. These various devices continue to generate various types of data or large amounts of multimedia data. To handle this situation, a large amount of storage is required, and big data technology is required to process stored data and obtain accurate information. NAS (Network Attached Storage) or SAN (Storage Area Network) technology is typically used to build high-speed, high-capacity storage in a network-based environment. In this paper, we propose a method to construct a mass storage device using Network-DAS which is an extension technology of DAS (Direct Attached Storage). Benchmark experiments were performed to verify the scalability of the storage system with 76 HDD. Experimental results show that the proposed high performance mass storage system is scalable and reliable.

• The Journal of the Korea Contents Association
• /
• v.11 no.3
• /
• pp.26-35
• /
• 2011

In these days, internet environment are very quickly development as well on-line service have been using a online for the mission critical business around the world. As the amount of information to be processed by computers has recently been increased there has been cluster computing systems developed by connecting workstations server using high speed networks for high availability. but cluster computing technology are limited for a lot of IT resources. So, grid computing is an expanded technology of distributed computing technology to use low-cost and high-performance computing power in various fields. Although the purpose of Grid computing focuses on large-scale resource sharing, innovative applications, and in some case, high-performance orientation, it has been used as conventional distributed computing environment like clustered computer until now because grid middleware does not have common sharable information system. In order to use grid computing environment efficiently which consists of various grid middleware, it is necessary to have application-independent information system which can share information description and services, and expand them easily. This paper proposed new database architecture and load balancing for high availability through Grid technology.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.8
• /
• pp.1-11
• /
• 2021

This paper proposes a novel mechanism called Event-driven Uncore Frequency Scaler (eUFS) to improve the energy efficiency of the HPC systems. UFS exploits the hardware events such as LAPI (Last-level Cache Accesses Per Instructions) and CPI (Clock Cycles Per Instruction) to dynamically adjusts the uncore frequency. Hardware events are collected at a reference time period, and the target uncore frequency is determined using the collected event and the previous uncore frequency. Experiments with the NPB benchmarks demonstrate that the eUFS reduces the energy consumption by 6% on average for class C and D NPB benchmarks while it only increases the execution time by 2% on average.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.170-175
• /
• 2022

Reactive failure management techniques are required to mitigate the impact of errors in high performance computing. Checkpoint is the standard recovery technique for coping with errors. An application employing checkpoints periodically saves its state, so that when an error occurs while some task is executing, the application is rolled back to its last checkpointed task and resumes execution from that task onward. In this paper, assuming the time-to-errors are independent each other and generally distributed, we analyze the checkpointing model with instantaneous error detection. The conventional assumption that two or more errors do not take place between two consecutive checkpoints is removed. Given the checkpointing time, down-time, and recovery time, we derive the reliability of the checkpointing model. When the time-to-error follows an exponential distribution, we obtain the optimal checkpointing interval to achieve the maximum reliability.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.251-260
• /
• 2002

The parallel/distributed processing with a lot of the idle hosts on the web has the high coot-performance ratio for large-scale applications. It's processing has to show the solutions for unpredictable status such as heterogeneity of hosts, variability of hosts, autonomy of hosts, the supporting performance continuously, and the number of hosts which are participated in computation and so on. In this paper, we propose the strategy of adaptive tack reallocation based on performance the host job processing, spread out geographically Also, It shows the scheme of dynamic host management with dynamic environment, which is changed by lots of hosts on the web during parallel processing for large-scale applications. This paper implements the PDSWeb (Parallel/Distributed Scheme on Web) system, evaluates and applies It to the generation of rendering image with highly intensive computation. The results are showed that the adaptive task reallocation with the variation of hosts has been increased up to maximum 90% and the improvement in performance according to add/delete of hosts.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.173-180
• /
• 2022

Approximate Computing is a promising method for designing hardware-efficient computing systems. Approximate multiplication is one of key operations used in approximate computing methods for high performance and low power computing. An approximate 4-2 compressor can implement hardware-efficient circuits for approximate multiplication. In this paper, we propose an approximate multiplier with low area and low power characteristics. The proposed approximate multiplier architecture is segmented into three portions; an exact region, an approximate region, and a constant correction region. Partial product reduction in the approximation region are simplified using a new 4:2 approximate compressor, and the error due to approximation is compensated using a simple error correction scheme. Constant correction region uses a constant calculated with probabilistic analysis for reducing error. Experimental results of 8×8 multiplier show that the proposed design requires less area, and consumes less power than conventional 4-2 compressor-based approximate multiplier.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.1
• /
• pp.9-22
• /
• 2022

Virtual Reality(VR) technology is an interface technology that is actively used in various audio-visual-based applications by showing users a virtual world composed of computer graphics. Since VR-based applications are graphic processing-based applications, expensive computing devices equipped with Graphics Processing Unit(GPU) are essential for graphic processing. This incurs a cost burden on VR application users for maintaining and managing computing devices, and as one of the solutions to this, a method of operating services in cloud environments is being used. This paper proposes a performance management technique to address the problem of performance interference between containers owing to GPU resource competition in container-based high-performance cloud environments in which multiple containers share a single GPU. The proposed technique reduces performance deviation due to performance interference, helping provide uniform performance-based remote VR services for users. In addition, this paper verifies the efficiency of the proposed technique through experiments.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.4
• /
• pp.369-377
• /
• 2008

Today, Ethernet technology is rapidly developing to have a bandwidth of 10Gbps beyond 1Gbps. In such high-speed networks, the existing method that host CPU processes TCP/IP in the operating system causes numerous overheads. As a result of the overheads, user applications cannot get the enough computing power from the host CPU. To solve this problem, the TCP/IP Offload Engine(TOE) technology was emerged. TOE is a specialized NIC which processes the TCP/IP instead of the host CPU. In this paper, we implemented a high-performance, lightweight TCP/IP(HL-TCP) for the TOE and applied it to an embedded system. The HL-TCP supports existing fundamental TCP/IP functions; flow control, congestion control, retransmission, delayed ACK, processing out-of-order packets. And it was implemented to utilize Ethernet MAC's hardware features such as TCP segmentation offload(TSO), checksum offload(CSO) and interrupt coalescing. Also we eliminated the copy overhead from the host memory to the NIC memory when sending data and we implemented an efficient DMA mechanism for the TCP retransmission. The TOE using the HL-TCP has the CPU utilization of less than 6% and the bandwidth of 453Mbps.

• KIISE Transactions on Computing Practices
• /
• v.22 no.10
• /
• pp.488-496
• /
• 2016

In this paper, we introduce a virtual reality interface for scientific visualization applications. Our VR interface is based on an open-source framework called VR Juggler. Although VR Juggler has its own advantages, it lacks some of the important functionalities needed for practical applications - event handling, synchronization and data sharing among cluster nodes, to name a few. We explain how these issues are resolved while developing the VR interface. Also, a new interface with a smart device, which replaces the virtual reality input device, is introduced. Finally, system usability test results are provided to prove the effectiveness of the proposed interfaces.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.3
• /
• pp.11-16
• /
• 2004

Recently, collaborative works are increased more and more over the distributed heterogeneous computing environments. The availability of high-speed wide-area networks has also enabled collaborative multimedia applications such as video conferencing, distributed interactive simulation and collaborative visualization. Distributed high performance computing and collaborative multimedia applications, it is extremely important to efficiently perform group communication over a heterogeneous network. Typical group communication patterns are broadcast and Multicast. Heuristic algorithms such as FEF, ECEF, look-ahead make up the message transmission tree for the broadcast and multicast over the distributed heterogeneous systems. But, there are some shortcomings because these select the optimal solution at each step, it may not be reached to the global optimum In this paper, we propose a new heuristic algerian that constructs tree for efficiently collective communication over the previous heterogeneous communication model which has heterogenity in both node and network. The previous heuristic algorithms my result in a locally optimal solution, so we present more reasonable and available criterion for choosing edge. Through the performance evaluation over the various communication cost, improved heuristic algorithm we proposed have less completion time than previous algorithms have, especially less time complexity than look-ahead approach.