• Journal of KIISE:Computer Systems and Theory
• /
• v.32 no.5
• /
• pp.232-242
• /
• 2005

This paper concerns automatic hardware synthesis from data flow graph (DFG) specification for fast HW/SW cosynthesis. A node in BFG represents a coarse grain block such as FIR and DCT and a port in a block may consume multiple data samples per invocation, which distinguishes our approach from behavioral synthesis and complicates the problem. In the presented design methodology, a dataflow graph with specified algorithm can be mapped to various hardware structures according to the resource allocation and schedule information. This simplifies the management of the area/performance tradeoff in hardware design and widens the design space of hardware implementation of a dataflow graph compared with the previous approaches. Through experiments with some examples, the usefulness of the proposed technique is demonstrated.

• The KIPS Transactions:PartA
• /
• v.15A no.6
• /
• pp.335-344
• /
• 2008

We propose an adaptive garbage collection technique for hybrid flash memory which has both SLC and MLC. Since SLC area is fast and MLC area has low cost, the proposed scheme utilizes the SLC area as log buffer and the MLC area as data block. Considering the high write cost of MLC flash, the garbage collection for the SLC log buffer moves a page into the MLC data block only when the page is cold or the page migration invokes a small cost. The other pages are moved within the SLC log buffer. Also it adjusts the parameter values which determine the operation of garbage collection adaptively considering I/O pattern. From the experiments, we can know that the proposed scheme provides better performance compared with the previous flash management schemes for the hybrid flash and finds the parameter values of garbage collection close to the optimal values.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.49-54
• /
• 2022

With the recent widespread adoption of general-purpose GPUs (GPGPUs) in cloud systems, maximizing the resource utilization through multitasking in GPGPU has become an important issue. In this article, we show that resource allocation based on the workload classification of computing-bound and memory-bound is not sufficient with respect to resource utilization, and present a new thread block scheduling policy for GPGPU that makes use of fine-grained resource utilizations of each workload. Unlike previous approaches, the proposed policy reduces scheduling overhead by separating profiling and scheduling, and maximizes resource utilizations by co-locating workloads with different bottleneck resources. Through simulations under various virtual machine scenarios, we show that the proposed policy improves the GPGPU throughput by 130.6% on average and up to 161.4%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.6
• /
• pp.1305-1317
• /
• 2019

Although encryption for data protection is essential in the big data platform environment of public institutions and corporations, much performance verification studies on encryption algorithms considering actual big data workloads have not been conducted. In this paper, we analyzed the performance change of AES, ARIA, and 3DES for each of six workloads of big data by adding data and nodes in MongoDB environment. This enables us to identify the optimal block-based cryptographic algorithm for each workload in the big data platform environment, and test the performance of MongoDB by testing various workloads in data and node configurations using the NoSQL Database Benchmark (YCSB). We propose an optimized architecture that takes into account.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.6
• /
• pp.95-107
• /
• 2008

In this paper, we propose a fast mode decision method for temporal and spatial scalability to reduce computational complexity of mode decision that used to be computationally one of the most intensive processes of the MPEG-4 AVC|H.264 SE(Scalable Extension) encoding. For temporal scalability, we propose an early skip method and MHM(mode history map) method. The early skip method confines macroblock modes of backward and forward frames within selected a few candidates. The MHM method utilizes stored information of frames inside a GOP of lower levels for the decision of MHM at higher level. For the spatial scalability, we propose the method that uses a candidate mode according to the MHM method and adds the BL_mode as candidates. The proposed scheme reduces the number of candidate modes to reduce computational complexity in mode decision. The proposed scheme reduces total encoding time by about 52% for temporal scalability and 47% for spatial scalability without significant loss of RD performance.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.8
• /
• pp.102-116
• /
• 2009

There have been lots of researches for H.264/AVC performance enhancement on a multi-core processor. The enhancement has been performed through parallelization methods. Parallelization methods can be classified into a task-level parallelization method and a data level parallelization method. A task-level parallelization method for H.264/AVC decoder is implemented by dividing H.264/AVC decoder algorithms into pipeline stages. However, it is not suitable for complex and large bitstreams due to poor load-balancing. Considering load-balancing and performance scalability, we propose a horizontal data level parallelization method for H.264/AVC decoder in such a way that threads are assigned to macroblock lines. We develop a mathematical performance expectation model for the proposed parallelization methods. For evaluation of the mathematical performance expectation, we measured the performance with JM 13.2 reference software on ARM11 MPCore Evaluation Board. The cycle-accurate measurement with SoCDesigner Co-verification Environment showed that expected performance and performance scalability of the proposed parallelization method was accurate in relatively high level

• Journal of the Korean Institute of Landscape Architecture
• /
• v.42 no.2
• /
• pp.1-9
• /
• 2014

This study is the renovation plan for BIFF(Busan International Film Festival) street themed movies, which is an internationally known film festival. The aim of this is the regional economic vitalization. The first step of the plan sets up position through the case study of the Theme Street. The theme for each road space is related with movies and realizes the unificative images for each road. Images between the Busan Cinema Center and BUSAN MARINA are introduction of the road and the subject of this road is greeting with movies. The history of movies are printed on the pavement, some sections are made with red blocks for recollecting the red carpet. The next section from the BUSAN MARINA to MARINE CITY is set up being close with movies. In this section, sculptures of filmmaking and theme benches are installed for indirect experiences. The theme from MARINE CITY to DONGBAEK ISLANDS is playing and enjoying with movies. It is made more fun with the installation of super graphic and trick art. The theme from DONGBAEK ISLANDS to HAEUNDAE is farewell with movies. It is expressed by music on pavement and musical fountain. The last section in the theme road from HAEUNDAE to MOONTAN ROAD shows the concept memories after farewell. It is a half way to Moonten Road. The Milky Way pavement and Milky Way square are made by installing the optical fibers.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.6C
• /
• pp.626-633
• /
• 2009

ISO/IEC/WD 24753 defines new modem specifications for a long-range RFID communications and application protocol for a sensor tag system. According to the standard, the frequency offset of the tag is 4%. In general wireless communications systems, it is known that a coherent receiver is superior to a non-coherent receiver. However, if the frequency offset is large, it is difficult to restore the original data accurately with a coherent receiver, and the performance of a coherent receiver is easily degraded. In this paper, a non-coherent receiver structure is adopted to solve the frequency offset problem of long-range RFID communications. We designed a frequency estimation block to find an optimal frequency from the received signal with 4% frequency offset and proposed a start frame delimiter (SFD) detection algorithm to determine the start position of the payload. The frequency estimation block finds the optimal frequency from the received signal using 9-correlators. And the SFD detection block searches the received signal to find the start position of the payload with dual correlator. We implemented a long-range RFID reader with the proposed methods and evaluated its performance in a wired/wireless test network. The implemented long-range RFID reader showed more superior performance than the commercial RFID reader in terms of recognition range.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2003.09a
• /
• pp.81-81
• /
• 2003

목적 : 현재, 많은 병원이 방사선과 의료영상정보를 기존의 필름형태로 판독하고, 진료하는 방식에서 PACS 를 도입하여 디지털 형태로 영상을 전송, 저장, 검색, 판독하는 환경으로 변화하고 있다. 한편, PACS 가 가지는 가장 큰 제한점은 휴대성의 결핍이다. 본 연구는 이동형 장치가 가지는 호스트의 이동성 및 휴대성의 장점들을 살리면서, 무선 채널 용량의 한계, 무선 링크 사용이라는 제약점들을 감안하여 의료영상을 JPEG2000 영상압축 방식으로 부호화한 후 무선 환경을 고려한 전송 패킷의 크기를 결정하고자 하였으며, 무선 통신 중 발생되는 패킷 손실에 대응하기 위한 자동 오류 수정 기능도 함께 구현하고자하였다. 방법 : Window 2000 운영체계에서 의료영상을 로드하고, 데이터베이스화하며, 저장하고, 다른 네트워크와 접속, 제어가 가능한 PC급 서버를 구축하였다. 영상데이터는 무선망을 통해 전송하기 때문에 가장 높은 압축비율을 지원하면서 에너지 밀도가 높은 JPEG2000 알고리즘을 사용하여 영상을 압축하였다. 또한, 무선망 사용으로 인한 패킷 손실에 대비하여, 영상을 JPEG2000 방식으로 부호화한 후 각 블록단위로 전송하였다. 결과 : PDA에서 JPEG2000 영상을 복호화 하는데 걸리는 시간은 256$\times$256 크기의 MR 뇌영상의 경우 바로 확인할 수 있었지만, 800$\times$790 크기의 CR 흉부 영상의 경우 약 5 초 정도의 시간이 걸렸다. CDMA 1X(Code Division Multiple Access 1st Generation) 모듈을 사용하여 영상을 전송하는 경우, 256 byte/see 정도에서는 안정된 전송 결과를 보여주었고, 1 Kbyte/see 정도의 전송의 경우 중간 중간에 패킷이 손실되는 결과를 관찰할 수 있었다. 반면 무선 랜의 경우 이보다 더 큰 패킷을 전송하더라도 문제점은 발견되지 않았다. 결론 : 현재의 PACS는 유선과 무선사이의 인터페이스의 부재로 인해 유무선 연동이 되지 못하고 있다. 따라서 이동형 JPEG2000 영상 뷰어는 PACS가 가지는 문제점인 휴대성을 보완하기 위하여 개발되었다. 또한 무선망이 가지는 데이터 손실에 대하여서도 허용할 수 있는 범위에서 재전송을 가능하게 함으로서 약한 연결성을 보완하였다. 본 JPEG2000 영상 뷰어 시스템은 기존 유선상의 PACS와 이동형 장치간에 유기적인 인터페이스 역할을 하리라 기대된다.

• Magazine of the Korea Concrete Institute
• /
• v.6 no.1
• /
• pp.89-100
• /
• 1994

The current test procedure for transfer length, which determine transfer length by measuring concrete strain, has an actual bond stress state in the prestressed pretensioned member : however, it is difficult to determine the bond properties of maximum bond stress and bond stiffness with this method. It is also difficult for design engineer to understand and select a correct safety criterion from the widely distributed results of such a ransfer test alone. An alternative testing procedure is provided here to determine the bond properties without measuring the concrete strain. In this test the bond stress is measured directly by creating a similar boundary condition within the transfer length in a real beam during the transfer of prestressing force. The prestressing force was released step by step by step from the unloading side. The release of force induces a swelling of the strand at the unloading side of concrete block, bonding force in the block, and a bond slip of the strand toward the other side of the block. Two center-hole load cells are used to record the end loads until the point of general bond slip(maximum bond stress). It is suggested that this test procedure be performed with the ordinary transfer test when determining the transfer length in a prestressed, pretensioned concrete beam.