• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.337-340
• /
• 2014

클라우드 시스템이 큰 이슈로 떠오르면서 그 기반이 되는 분산 파일 시스템에 관한 연구가 계속되고 있다. 최근 제안된 분산파일 시스템은 대부분 확장 가능하며 신뢰성이 있는 시스템으로 구성되어 있으며 내고장성(Fault tolerance)과 높은 가용성을 위해 데이터 복제 기법을 사용하며 하둡 분산 파일 시스템에서는 블락의 복제수를 기본3개로 지정한다. 그러나 이 정책은 복제수가 많아지면 많아질수록 가용성은 높아지지만 스토리지 또한 증가한다는 단점이 있다. 본 논문에선 이러한 문제점을 해결하기 위해 최소한의 블락 복제수와 복제된 블락을 효율적으로 배치하여 더 좋은 성능과 부하분산(Load Balancing)하기 위한 기법을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.16-18
• /
• 2021

최근 베어메탈 시스템에 스케줄러 시스템 기반 인프라 시스템에서 벗어나 사용자 맞춤형 클라우드 인프라가 관심을 받고 있다. KI Cloud 플랫폼은 고성능 컴퓨팅(HPC) 사용자를 위한 클라우드 기반 IaaS, PaaS 통합 플랫폼이다. 본 논문에서는 KI Cloud를 위한 데이터 저장소 설계 및 구현에 대해 기술한다. 오픈 소스 Ceph 분산 스토리지 시스템을 기반으로 구축하였고, 프로젝트의 목적에 맞게 블록 스토리지, 오브젝트 스토리지, 파일 스토리지 방식으로 구현하였다. Openstack 기반의 서비스와 Kubernetes 기반의 서비스가 하나의 스토리지를 통해 다양한 타입의 서비스를 제공할 수 있도록 개발하였다.

• Journal of Korean Society of Archives and Records Management
• /
• v.19 no.3
• /
• pp.271-292
• /
• 2019

Recently, the central administrative agency changed its business management system to cloud-based On-nara 2.0. To transfer and manage the records of the cloud business management system, the National Archives Service has developed and distributed a cloud-based records management system. It serves as an opportunity to maximize the benefits of cloud computing and redesign the records management to be more effective and efficient. The process and method of electronic record management can be transformed through digital technologies. First, we can change the transfer method for electronic records. When the business and the records management systems share the same cloud storage, it is possible to transfer the content files between the two systems without moving the contents files physically, thus copying only the metadata and reducing the cost and the risk of integrity damage. Second, the strategy for allocating storage space for contents can be conceived. Assuming that the cloud storage is shared by the business and the record management systems, it is advantageous to distinguish the storage location based on the retention period of the content files. Third, systems that access content files, such as records creation, records management, and information disclosure systems, can share the cloud storage and minimize the duplication of content files.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.2
• /
• pp.357-367
• /
• 2018

As information technology of modern society develops, various malicious codes with the purpose of seizing or destroying important system information are developing together. Among them, ransomware is a typical malicious code that prevents access to user's resources. Although researches on detecting ransomware performing encryption have been conducted a lot in recent years, no additional methods have been proposed to recover damaged files after an attack. Also, because the similarity comparison technique was used without considering the repeated encryption, it is highly likely to be recognized as a normal behavior. Therefore, this paper implements a filter driver to control the file system and performs a similarity comparison method that is verified based on the analysis of the encryption pattern of the ransomware. We propose a system to detect the malicious process of the accessed process and recover the damaged file based on the cloud storage.

• The Journal of the Korea Contents Association
• /
• v.18 no.4
• /
• pp.178-185
• /
• 2018

Recently, demands for computer systems with multicore CPUs and high-performance flash-based storage devices (i.e., flash SSD) have rapidly grown in cloud computing, surer-computing, and enterprise storage/database systems. Journaling file systems running on high-performance systems do not exploit the full I/O bandwidth of high-performance SSDs. In this article, we evaluate and analyze the performance of the Linux EXT4 file system with high-performance SSDs and multicore CPUs. The system used in this study has 72 cores and Intel NVMe SSD, and the flash SSD has performance up to 2800/1900 MB/s for sequential read/write operations. Our experimental results show that checkpointing in the EXT4 file system is a major overhead. Furthermore, we optimize the checkpointing procedure and our optimized EXT4 file system shows up to 92% better performance than the original EXT4 file system.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.3
• /
• pp.521-530
• /
• 2017

In this paper, we propose a protection solution against intelligent Ransomware attacks by encrypting not only source files but also backup files of external storage. The system is designed to automatically back up to the cloud server at the time of file creation to perform monitoring and blocking in case a specific process affects the original file. When client creates or saves a file, both process identifiers, parent process identifiers, and executable file hash values are compared and protected by the whitelist. The file format that is changed by another process is monitored and blocked to prevent from suspicious behavior. By applying the system proposed in this paper, it is possible to protect against damage caused by the modification or deletion of files by Ransomware.

• The Journal of the Korea Contents Association
• /
• v.15 no.5
• /
• pp.570-581
• /
• 2015

In recent years, many institutions predict that cloud services and big data will be popular IT trends in the near future. A number of leading IT vendors are focusing on practical solutions and services for cloud and big data. In addition, cloud has the advantage of unrestricted in selecting resources for business model based on a variety of internet-based technologies which is the reason that provisioning and virtualization technologies for active resource expansion has been attracting attention as a leading technology above all the other technologies. Big data took data prediction model to another level by providing the base for the analysis of unstructured data that could not have been analyzed in the past. Since what cloud services and big data have in common is the services and analysis based on mass amount of data, efficient operation and designing of mass data has become a critical issue from the early stage of development. Thus, in this paper, I would like to establish data processing architecture based on technological requirements of mass data for cloud and big data services. Particularly, I would like to introduce requirements that must be met in order for distributed file system to engage in cloud computing, and efficient compression technology requirements of mass data for big data and cloud computing in terms of cost-saving, as well as technological requirements of open-source-based system such as Hadoop eco system distributed file system and memory database that are available in cloud computing.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.874-877
• /
• 2010

웹 서비스 혹은 클라우드 컴퓨팅 서비스로써 대용량의 멀티미디어 데이터에 대한 내용 기반 검색을 지원하기 위하여, 분산 고차원 색인 구조에 대한 연구가 활발하게 이뤄지고 있다. 이러한 고차원 데이터의 색인 구조에 대한 연구와 달리, 고차원 색인 데이터의 저장 및 관리에 대한 연구는 거의 전무한 것이 현실이다. 지금껏 대부분의 색인 데이터는 빠른 접근을 위하여 트랜잭션 관리 및 데이터의 복구를 제공하지 않은 파일 시스템에서 관리되어 왔다. 그러나, 파일 시스템에 저장된 색인 데이터는 이를 서비스하는 노드의 장애 발생 시에 일부 혹은 전체 데이터에 대한 검색이 이뤄지지 않는 문제점이 있다. 서비스의 가용성 여부가 중요한 요소인 웹 서비스와 클라우드 컴퓨팅 서비스를 위하여, 본 논문에서는 고차원 색인 데이터를 데이터베이스를 통해 관리하여, 안정성 및 가용성을 보장하면서, 고차원 데이터의 색인 및 검색의 성능을 보장하는 방법을 제안하고자 한다.

• Journal of Internet Computing and Services
• /
• v.14 no.1
• /
• pp.53-62
• /
• 2013

It is common among users to have multiple computing devices as well as to access their files or do work at different locations. To achieve file consistency as well as mobility in this scenario, an efficient approach for workspace synchronization should be used. However, file synchronization alone cannot guarantee the mobility of work environment which allows activities to be resumed at any place and time. This paper proposes a ubiquitous synchronization approach which provides cloud-based access to a user's workspace. Efficient synchronization is achieved by combining session monitoring with file system management. Experimental results show that the proposed mechanism outperforms Cloud Master-replica Synchronization in terms of number of I/O operations, CPU utilization, as well as the average and maximum latencies in responding to client requests.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.145-153
• /
• 2023

In a P2P-based distributed storage system where peers frequently join and leave, it is essential to guarantee not only data availability but also playback quality comparable to that provided by local storage devices when playing back video files with real-time constraints. In addition, cloud storage services based on distributed storage systems provide each user with the functionality to share their files with other users, so when multiple users request playback of the same video file at the same time, all playback should be supported seamlessly in real time. Therefore, in this paper, we propose a scheme that process multiple simultaneous playback requests for each video file in real time as well as data availability in a P2P-based distributed storage system that supports file sharing. This scheme can support real-time simultaneous playback and efficiently use storage space by adjusting the amount of redundant data encoded through erasure coding according to the number of concurrent playback requests for each video file.