• Journal of Korea Society of Digital Industry and Information Management
• /
• v.13 no.4
• /
• pp.153-162
• /
• 2017

Advances in communication technologies and the decreasing cost of computers have made distributed computer systems an attractive alternative for satisfying the information needs of large organizations. This paper presents a distributed algorithm for performance improvement through load balancing and file migration in distributed systems. We employed a sender initiated strategy for task migration and used learning automata with several internal states for file migration. A task can be migrated according to the load information of a computer. A file is migrated to the destination processor when it is in the right boundary state. We also described an analytical model for load balancing with file migration to verify the proposed algorithm. Analytical and simulation results show that our algorithm is very well-suited for distributed system environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.709-735
• /
• 2010

The reciprocal virtue of peer-to-peer networking has stimulated an explosion of peer population and service capacity, ensuring rapid content distribution in peer-to-peer networks. Critical issues such as peer churn, free riding, and skewed workload significantly affect performance results such as service agility, fairness, and resource utilization. To resolve these problems systematically, this study proposes a peer assignment scheme that supports fair peer-to-peer file sharing applications. The proposed scheme exploits the peer duality of both server-oriented peer capacity and client-oriented peer contribution. Accordingly, the system server can prioritize download requests and appropriately assign server peers to uploading file objects. Several functional extensions, including peer substitution and elimination, bandwidth adjustment, and distributed modification, help cope with subtle situations of service starvation and download blocking, and hence make the system design robust and amenable. Simulation results show this design is examined under both centralized and distributed peer-to-peer environments. Performance results confirm that the proposed mechanisms are simple but effective in maintaining service agility and fairness, without loss of overall service capacity in peer-to-peer files sharing systems.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.16 no.2
• /
• pp.11-25
• /
• 2020

This study proposes a method to manipulate fragmentation of disks by arbitrarily allocating and releasing the status of a disk cluster in the NTFS file system. This method allows experiments to be performed in several studies related to fragmentation problems on disk cluster. Typical applicable research examples include testing the performance of disk defragmentation tools according to the state of fragmentation, establishing an experimental environment for fragmented file carving methods for digital forensics, setting up cluster fragmentation for testing the robustness of data hiding methods within directory indexes, and testing the file system's disk allocation methods according to the various version of Windows. This method suggests how a single file occupies a cluster and presents an algorithm with a flowchart. It raises three tricky problems to solve the method, and we propose solutions to the problems. Experiments for allocating the disk cluster to be fragmented to the maximum extent possible, it then performs a disk defragmentation experiment to prove the proposed method is effective.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.31A no.10
• /
• pp.21-30
• /
• 1994

We discussed the Distributed Control System's design on preface and analyzed time of the real-time communication by using designed system. The DCS proposed in this thesis was implemented to network file system having recovery advantage and shared memory method to access file system of a Remote Station with ease. Also, this system minimized the network delay-time by using the real-time VME147 board. In implemented DCS, the performance analysis of real-time process of a Remote Station was done to get the total time for reak-tune communication from a Remote Station to the Central Station after terminating of process. For the analysis of system performance, we experiented by three steps. Firstly, we measuredthe processing the of LOOP function that real-time CPU convertes to-2,500~10.000 values from the input data of the Analog Interface Card. Secondly, we measured the processing time of the LOGIC function and the LOOP function. Lastly, we measured total processing time for communication from a Remote Station to the Centrol Station.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.275-278
• /
• 1996

We propose dynamic file allocation method in distributed database management system with changing access patterns. There are a lot of studies on file allocation problem in D-DBMS, and those studies deal with off-line analysis and optimization. Those works are well for systems with static database access patterns, but are inadequate for systems that have changing access patterns. In these systems, dynamic file allocation along with access pattern is more proper. In advance, Brunstrom et al. studied on this area, but they dealt a extremely simplified model. So, we make more practical models to simulate real system. In these models, many factors that were disregard in the advance study are considered. These models are composed with the non-replication system and the replication system. In addition to, we deal with CPU workload balancing in such system in order to improve performance of systems. Our methodology is very simple and realistic, therefore we think that it will give a lot of improvement in D-DBMS with changing access pattern.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.9
• /
• pp.2827-2835
• /
• 2000

In recent years, much attention has been focused on improving I/O devices' processing speed which is essential in such large data processing areas as multimedia data processing. And studies on high-performance parallel file systems are considered to be one of such efforts. In this paper, an efficient disk allocation scheme is proposed for high-performance parallel file systems. In other words, the concept of a parallel disk file's parallelism is defined using data declustering characteristic of a given parallel file. With the concept, an efficient disk allocation scheme is proposed which calculates the appropriate degree of data declustering on disks for each parallel file in order to obtain the maximum throughput when more than one parallel file is used at the same time. Since, calculation for obtaining the maximum throughput is too complex as the number of parallel files increases, an approximate disk allocation algorithm is also proposed in this paper. The approximate algorithm is very simple and especially provides very good results when I/O workload is high. In addition, it has shown that the approximate algorithm provides the optimal disk allocation for the maximum throughput when the arrival rate of I/O requests is infinite.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.1-8
• /
• 2022

As an efficient file system for SSD(Solid State Drive), F2FS is employed in the kernel of Linux operating system. F2FS applies various methods to improve performance by reflecting the characteristics of flash memory. One of them is improvement of the index structure that contains addresses of data blocks for each file. This paper presents a method for further improving performance by modifying the index structure of F2FS. F2FS manages all index blocks as logical numbers, and an address mapping table is used to find the physical block addresses of index blocks on flash memory. This paper shows performance improvement by applying logical numbers to the last level index blocks only. The count of mapping table search for a data block access is reduced to 1~2 from 1~4.

• Journal of Communications and Networks
• /
• v.15 no.2
• /
• pp.228-239
• /
• 2013

Peer-to-peer (P2P) systems have proved the most effective and popular file sharing applications in recent years. Previous studies mainly focused on equal service and differentiated service strategies when peers have no initial data before their downloads. For an upload-constrained P2P file sharing system, we model both the equal service process and the differentiated service process when the initial data distribution of peers satisfies some special conditions. Moreover, we show how to minimize the time required to distribute the file to any number of peers. The proposed fluid-based models can reveal the intrinsic relations among the initial data amount, the peer set size, and the minimum last finish time. The closed-form expressions derived from the extended models can closely approximate chunk-based models and systems, especially for relatively large files. As an application of the extended models, we show how to provide differentiated service efficiently to multiple peer sets. Since no limits are imposed on the upload bandwidth of peers or the size of each peer set, we believe that our analytic process and the results achieved can provide not only fundamental insights into bandwidth allocation and data scheduling but also a helpful reference for both improving system performance and building an effective incentive mechanism for P2P file sharing systems.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1221-1230
• /
• 1997

In this study, the performance of a parallel file system(N-PFS), which is inplemented using conventional disks as disk arrays on a Workstation Cluster, is analyzed by using analytical method and adtual values in experiments.N-PFS can be used as high-performance file sever in small-scale server systems and effciently pro-cess massive data I/Os such as multimedia and scientifid data. In this paper, an analytical model was suggested and the correctness of the suggested was verified by analyzing the experimental values on a system.The result of the appropriate stping unit for processing massive data of the Workstation Cluster with 8 disks is 64-128Kbytes and the maximum throughput on it is 15.8 Mbytes/ses.In addition, the performance of parallel file system on massive data is bounded by the time required to copy data between buffers.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.1
• /
• pp.86-98
• /
• 2019

Recently, the trend of building container-based PaaS (Platform-as-a-Service) is expanding. Container-based platform technology has been a core technology for realizing a PaaS. Containers have lower operating overhead than virtual machines, so hundreds or thousands of containers can be run on a single physical machine. However, recording and monitoring the storage logs for a large number of containers running in cloud computing environment occurs significant overhead. This work has identified two problems that occur when detecting a file system change event of a container running in a cloud computing environment. This work also proposes a system for container file system event detection in the environment by solving the problem. In the performance evaluation, this work performed three experiments on the performance of the proposed system. It has been experimentally proved that the proposed monitoring system has only a very small effect on the CPU, memory read and write, and disk read and write speeds of the container.