• 정보처리학회논문지A
• /
• 제14A권7호
• /
• pp.443-450
• /
• 2007

대용량 무선 인터넷 프록시 캐시 서버 클러스터에서는 성능 및 저장 공간의 확장성이 중요하게 되었다. 여기에서 성능의 확장성은 캐시 서버를 추가함에 따라 클러스터 성능이 선형적으로 증가함을 의미하고 저장 공간의 확장성은 캐시 데이터가 서버들에게 분할 저장되어 있어서 캐시 서버의 수에 상관없이 캐시 데이터를 저장하는 클러스터안의 공간의 합은 일정함을 의미한다. 대용량 서버 클러스터에서 많이 사용되는 라운드로빈 기반 부하분산 방법은 성능의 확장성은 보장되지만, 요청 URL 데이터가 모든 서버에 저장되어야 하므로 저장 공간의 확장성이 없는 단점을 가진다. 해싱기반 부하분산 방법은 모든 요청 URL 데이터가 서버들에 분할 저장되어 있어서 저장 공간의 확장성을 가진다. 그러나 그 방법은 사용자의 요청 패턴 불균형 또는 특정서버로의 요청 집중(Hot-Spot) 시에 성능 확장성이 없다. 본 논문에서는 성능 및 저장 공간의 확장성을 보장하는 새로운 동적 해싱 부하분산기법을 제안한다. 제안된 기법에서는 주기적으로 과부하 상태의 캐시 서버에 할당된 요청들을 찾아서 다른 캐시서버로 동적으로 재할당한다. 제안된 방법을 16대의 컴퓨터를 사용하여 실험을 수행하였고, 실험 결과를 통해 제안된 방법이 기존 방법과는 달리 성능 및 저장 공간의 확장성을 보장함을 확인하였다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제12권1호
• /
• pp.66-74
• /
• 2012

In this paper, the impact of segregation layer density ($N_{DSL}$) and length ($L_{DSL}$) on scalability and analog/RF performance of dopant-segregated Schottky barrier (DSSB) SOI MOSFET has been investigated in sub-30 nm regime. It has been found that, although by increasing the $N_{DSL}$ the increased off-state leakage, short-channel effects and the parasitic capacitances limits the scalability, the reduced Schottky barrier width at source-to-channel interface improves the analog/RF figures of merit of this device. Moreover, although by reducing the $L_{DSL}$ the increased voltage drop across the underlap length reduces the drive current, the increased effective channel length improves the scalability of this device. Further, the gain-bandwidth product in a common-source amplifier based on optimized DSSB SOI MOSFET has improved by ~40% over an amplifier based on raised source/drain ultrathin-body SOI MOSFET. Thus, optimizing $N_{DSL}$ and $L_{DSL}$ of DSSB SOI MOSFET makes it a suitable candidate for future nanoscale analog/RF circuits.

• Journal of Computing Science and Engineering
• /
• 제6권4호
• /
• pp.294-309
• /
• 2012

In the age of multi-core and specialized accelerators in high performance computing (HPC) systems, it is critical to understand application characteristics and apply suitable optimizations in order to fully utilize advanced computing system. Often time, the process involves multiple stages of application performance diagnosis and a trial-and-error type of approach for optimization. In this study, a general guideline of performance optimization has been demonstrated with two class-representing applications. The main focuses are on node-level optimization and inter-node scalability improvement. While the number of optimization case studies is somewhat limited in this paper, the result provides insights into the systematic approach in HPC applications performance engineering.

• 디지털융복합연구
• /
• 제18권10호
• /
• pp.253-263
• /
• 2020

2015년 이더리움 탄생 이후 많은 블록체인 DApp들이 등장하였다. 그러나 블록체인 게임 크립토키티의 사례에서 블록체인 플랫폼의 성능 및 확장성의 한계 또한 드러났다. 이를 해결하기 위해 높은 TPS(Transaction per second)를 개발하였다는 블록체인 플랫폼이 등장하였지만 아직까지 실제 이를 테스트해볼 수 있는 환경 및 서비스는 등장하지 못했다. 이에 본 논문에서는 모든 게임 내용을 체인 상에 기재하여 의도적으로 과대화된 Transaction의 발생을 유발하는 게임을 설계하고 개발하였다. 개발된 게임은 총 682명의 사용자를 통해 총 630만개의 블록과 890만개의 Transaction을 발생시켜 기존 플랫폼 대비 개선된 성능과 확장성을 검증할 수 있었다. 또한 테스트 기간 내의 최대 TPS는 1,309로 측정되었다. 향후 다양한 블록체인 플랫폼들에서 본 논문에서 제시한 방법으로 블록 및 Transaction 생성 수와 최대 TPS를 측정한다면 각 블록체인 플랫폼의 실제 상용 환경에서의 성능과 확장성을 비교할 수 있을 것으로 기대된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권4호
• /
• pp.452-474
• /
• 2010

Formidable growth of Internet technologies has revealed challenging issues about its scale and performance evaluation. Modeling and simulation play a central role in the evaluation of the behavior and performance of the large-scale network systems. Large numbers of nodes affect simulation performance, simulation execution time and scalability in a weighty manner. Most of the existing simulators have numerous problems such as size, lack of system theoretic approach and complexity of modeled network. In this work, a scalable discrete-event modeling approach is described for studying networks' scalability and performance traits. Key fundamental attributes of Internet and its protocols are incorporated into a set of simulation models developed using the Discrete Event System Specification (DEVS) approach. Large-scale network models are simulated and evaluated to show the benefits of the developed network models and approaches.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2010년도 추계학술대회
• /
• pp.188-191
• /
• 2010

The study investigated the optimal configuration of SVC (Scalable Video Coding) to apply to the satellite broadcasting service, and compared the performance of the SVC with that of the AVC (Advanced Video Coding). To get the optimal configuration, we analyzed the optimal bit rate allocation between the layers and the optimal scalability which requires the least bit rate for the required PSNR for various kinds of contents using JSVM. As a result of investigation, we found that the optimal bit rate allocation occurs when the bit rate of the base layer is minimum, and the spatial scalability shows the best performance. The performance of SVC is similar to that of AVC for spatial scalability, but it depends on contents.

• Journal of Electrical Engineering and information Science
• /
• 제2권6호
• /
• pp.48-55
• /
• 1997

Advances in VLSI technology have brought us completely new design principles for the high-performance switching fabrics including ATM switches. From a practical point of view, port scalability of ATM switches emerges as an important issue while complexity and performance of the switches have been major issues in the switch design. In this paper, we propose a cost-effective approach to modular ATM switch design which provides the good scalability. Taking advantages of both time-division and space-division switch architectures, we propose a practically implementable large scale ATM switch architecture. We present a scalable shared buffer type switch for a building block and its expansion method. In our design, a large scale ATM switch is realized by interconnecting the proposed shared buffer switches in three stages. We also present an efficient control mechanism of the shared buffers, synchronization method for the switches in each stage, and a flow control between stages. It is believed that the proposed approach will have a significant impact on both improving the ATM switch performance and enhancing the scalability of the switch with a new cost-effective scheme for handling the traffic congestion. We show that the proposed ATM switch provides an excellent performance and that its cell delay characteristic is comparable to output queueing which provides the best performance in cell delay among known approaches.

• 정보처리학회논문지A
• /
• 제13A권2호
• /
• pp.137-146
• /
• 2006

최근 Chip-level MuitiThreading(CMT) 기술을 내장한 프로세서 들이 출시되면서 그들을 기반으로 하는 공유 메모리 다중 프로세서(SMP: Shared Memory Multiprocessor) 서버 또한 그 사용이 점점 더 보편화 되고있다. OpenMP는 그 사용의 효율성으로 인하여 SMP 시스템을 위한 응용 프로그램의 병렬화를 위한 표준이 되었다. 고성능 컴퓨팅(HPC: High Performance Computing) 응용프로그램 분야에서 더욱 더 빠른 컴퓨터의 처리 능력에 대한 요구가 증가함에 따라, OpenMP 지시어를 사용하여 병렬화된 HPC 응용 프로그램 들의 성능과 확장성을 높이는 일은 그 중요성이 점차 증대되고 있다. 본 논문에서는 CMT 기술을 내장한 대용량 SMP서버인 Sun Fire E25K에서 OpenMP 지시어를 사용하여 병렬화된 HPC 응용 프로그램 들의 suite인 SPEC OMPL(OpenMP를 위한 표준 벤치마크 suite)의 성능과 확장성에 관해 연구했다. 본 논문에서는 또한 SPEC OMPL에 대한 CMT 기술의 효능을 평가하였다.

• Journal of Computing Science and Engineering
• /
• 제7권1호
• /
• pp.1-20
• /
• 2013

Over the past generation, data warehousing and online analytical processing (OLAP) applications have become the cornerstone of contemporary decision support environments. Typically, OLAP servers are implemented on top of either proprietary array-based storage engines (MOLAP) or as extensions to conventional relational DBMSs (ROLAP). While MOLAP systems do indeed provide impressive performance on common analytics queries, they tend to have limited scalability. Conversely, ROLAP's table oriented model scales quite nicely, but offers mediocre performance at best relative to the MOLAP systems. In this paper, we describe a storage and indexing framework that aims to provide both MOLAP like performance and ROLAP like scalability by essentially combining some of the best features from both. Based upon a combination of R-trees and bitmap indexes, the storage engine has been integrated with a robust OLAP query engine prototype that is able to fully exploit the efficiency of the proposed storage model. Specifically, it utilizes an OLAP algebra coupled with a domain specific query optimizer, to map user queries directly to the storage and indexing framework. Experimental results demonstrate that not only does the design improve upon more naive approaches, but that it does indeed offer the potential to optimize both query performance and scalability.

• 인터넷정보학회논문지
• /
• 제5권6호
• /
• pp.1-10
• /
• 2004

A web robot is a software that has abilities of tracking and collecting web documents on the Internet(l), The performance scalability of recent web robots reached the limit CIS the number of web documents on the internet has increased sharply as the rapid growth of the Internet continues, Accordingly, it is strongly demanded to study on the performance scalability in searching and collecting documents on the web. 'Design of web robot based on Multi-Agent to speed up documents collection ' rather than 'Sequentially executing Web Robot based on the existing Fork-Join method' and the results of analysis on its performance scalability is presented in the thesis, For collection speedup, a Multi-Agent based web robot performs the independent process for inactive URL ('Dead-links' URL), which is caused by overloaded web documents, temporary network or web-server disturbance, after dividing them into each agent. The agents consist of four component; Loader, Extractor, Active URL Scanner and inactive URL Scanner. The thesis models a Multi-Agent based web robot based on 'Amdahl's Law' to speed up documents collection, introduces a numerical formula for collection speedup, and verifies its performance improvement by comparing data from the formula with data from experiments based on the formula. Moreover, 'Dynamic URL Partition algorithm' is introduced and realized to minimize the workload of the web server by maximizing a interval of the web server which can be a collection target.