• The Transactions of the Korea Information Processing Society
• /
• v.7 no.4
• /
• pp.1201-1210
• /
• 2000

In order to tolerate faults which may occur during the execution of distributed tasks in high-performance parallel computer systems, tasks are duplicated on different processors. In this paper, by utilizing the task duplication based scheduling algorithm, a new task scheduling algorithm which duplicates each task on more than two different processors with the minimum schedule length is presented, and the number of processors required for the duplication is analyzed with the ratio of communication cost to computation time and the workload of the system. A simulation with various task graphs reveals that the number of processors required for the full-duplex fault-tolerant task scheduling with the obtainable minimum schedule length increases about 30% to 75% when compared with that of the task duplication based scheduling algorithm.

• Smart Media Journal
• /
• v.2 no.3
• /
• pp.47-53
• /
• 2013

Recently convergence of IT in logistics system as a typical application RFID/NFC technology is being used, such as, according to the distribution of the flow is generated by a lot of big data. The Hadoop distributed system to collect data items produced by the parallel processing capabilities of logistics information and logistics information for the record management can create. Hadoop system to support the design and development of prototypes were approaching the possibility of its utilization.

• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.137-140
• /
• 2019

교통 시뮬레이션 시스템은 현실 세계의 교통 및 차량 관련 데이터를 기반으로 미래의 차량 움직임을 예측하는 프로그램으로, 다양한 교통문제를 해결을 위한 도구가 될 수 있다. 시뮬레이션 스케일을 전국단위로 확장하기 위해서 분산/병렬 시스템을 도입해야 하는데, 이 논문에서는 병렬/분산 과정에서 핵심이 되는 Open MPI 기반의 데이터 교환에 대한 방법을 제안하고자 한다. 공통된 하나의 커뮤니케이션 모듈을 기반으로 분산된 노드의 데이터 교환에 대한 문제를 해결하여 생산성을 높이고, 시뮬레이션 과정에서 소요되는 커뮤니케이션 타임을 줄여줄 것으로 예상된다.

• Annual Conference of KIPS
• /
• 2009.11a
• /
• pp.113-114
• /
• 2009

클러스터 시스템의 응용 분야가 다양화되고 복잡해짐에 따라, 대규모 클러스터 시스템을 보다 효율적으로 사용하기 위해서 실제 사용자의 이용 패턴을 예측할 수 있는 워크로드 분석의 필요성이 높아지고 있다. 이에 본 논문에서는 현재 가동중인 188개의 계산 노드, 3008개 CPU 자원을 보유한 대규모 클러스터 시스템에서 병렬 분산 파일 시스템에 대한 워크로드를 분석하였다.

• Annual Conference of KIPS
• /
• 2011.11a
• /
• pp.126-129
• /
• 2011

대용량 데이터의 분산/병렬 가시화를 위해서는 가시화 클라이언트와 서버 사이의 프로토콜이 필요하다. 기존 가시화 도구들은 개발 도구에 특화된 프로토콜을 사용하고 있으며, 이 때문에 클라이언트와 서버는 매우 tightly-coupled 되어 있다. 본 논문에서는 응용에 독립적인 분산/병렬 가시화를 위한 가시화 프로토콜을 설계하고 구현한다. 또한, 시변환 데이터의 효율적 가시화를 위해 animation을 구현할 수 있는 프리미티브를 설계하고 status machine으로 병렬 전송된 데이터들 간의 동기화를 구현한다. 이러한 응용 독립적 가시화 프로토콜을 도입함으로써 가시화는 병렬 분산 가시화를 수행하는 그리드의 서비스나 슈퍼컴퓨팅의 서비스로 확장될 수 있을 것이다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.3
• /
• pp.264-276
• /
• 2000

Recent advances of technologies make easy sharing various information and utilizing system resources on the Internet. Especially, code migration using applets of Java supports the distribution of programs on the web environment, and also browsers executing the applets guarantee the reliability of a migrated codes. In this paper, we describe the design and implementation of a web-based parallel processing system, which distributes migratable codes of a large job, makes the distributed codes to execute in parallel, and controls and gathers the results of each execution. The hosts participate in the computation reside on the Internet, spreaded out geographically, and the heterogeneity and the variability among them are severe. Thus, task allocation considering the performance differences and the adaptability to the severe variability are necessary. We present an adaptive task allocation algorithm applied to our system and the performance evaluation.

• Journal of Information Processing Systems
• /
• v.8 no.4
• /
• pp.555-566
• /
• 2012

The cloud environment makes it possible to analyze large data sets in a scalable computing infrastructure. In the bioinformatics field, the applications are composed of the complex workflow tasks, which require huge data storage as well as a computing-intensive parallel workload. Many approaches have been introduced in distributed solutions. However, they focus on static resource provisioning with a batch-processing scheme in a local computing farm and data storage. In the case of a large-scale workflow system, it is inevitable and valuable to outsource the entire or a part of their tasks to public clouds for reducing resource costs. The problems, however, occurred at the transfer time for huge dataset as well as there being an unbalanced completion time of different problem sizes. In this paper, we propose an adaptive resource-provisioning scheme that includes run-time data distribution and collection services for hiding the data transfer time. The proposed adaptive resource-provisioning scheme optimizes the allocation ratio of computing elements to the different datasets in order to minimize the total makespan under resource constraints. We conducted the experiments with a well-known sequence alignment algorithm and the results showed that the proposed scheme is efficient for the cloud environment.

• The KIPS Transactions:PartA
• /
• v.19A no.3
• /
• pp.129-138
• /
• 2012

In distributed memory architectures like clusters, each node stores a portion of data. How data is distributed across nodes influences the performance of such systems. The data distribution scheme is the strategy to distribute data across nodes and realize parallel data processing. Due to various reasons such as maintenance, scale up, upgrade, etc., the performance of nodes in a cluster can often become non-identical. In such clusters, data distribution without considering performance cannot efficiently distribute data on nodes. In this paper, we propose a new data distribution scheme based on the number of cores in nodes. We use the number of cores as the performance factor. In our data distribution scheme, each node is allocated an amount of data proportional to the number of cores in it. We implement our data distribution scheme using the Chapel language. To show our data distribution is effective in reducing the execution time of parallel applications, we implement Mandelbrot Set and ${\pi}$-Calculation programs with our data distribution scheme, and compare the execution times on a cluster. Based on experimental results on clusters of 8-core and 16-core nodes, we demonstrate that data distribution based on the number of cores can contribute to a reduction in the execution times of parallel programs on clusters.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.5
• /
• pp.257-264
• /
• 2002

In this paper, we give a detailed description of KDSM(KAIST Distributed Shared Memory) system. KDSM is implemented as a user-level library running on Linux 2.2.13, and TCP/IP is used for communication. KDSM uses page-based invalidation protocol, multiple-writer protocol, and supports HLRC(Home-based Lazy Release Consistency) memory consistency model. To evaluate performance of KDSM, we executed 4 scientific applications and compared the result to JLAJLA. The results showed that performance of KDSM almost equal to JIAJIA for 2 applications and performance of KDSM is better than JIAJIA for 2 applications.

• The KIPS Transactions:PartA
• /
• v.10A no.2
• /
• pp.123-130
• /
• 2003

This paper presents a new striped checkpointing scheme for serverless cluster computers, where the local disks are attached to the cluster nodes collectively form a distributed RAID with a single I/O space. Striping enables parallel I/O on the distributed disks and staggering avoids network bottleneck in the distributed RAID. We demonstrate how to reduce the checkpointing overhead and increase the availability by striping and staggering dynamically for communication intensive applications. Linpack HPC Benchamark and MPI programs are applied to these checkpointing schemes for performance evaluation on the 16-nodes cluster system. Benchmark results prove the benefits of the striped checkpointing scheme compare to the existing schemes, and these results are useful to design the efficient checkpointing scheme for fast rollback recovery from any single node failure in a cluster system.