• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1628-1631
• /
• 2015

웹을 통한 다수의 정보서비스와 전산자원을 공동으로 활용하는 경우에 있어서는, 구성자원에 대한 통합과 공동 활용이 상당히 중요한 요소이다. 일반적으로 웹 서비스를 위한 구성요소로서는 Web, WAS, 데이터베이스, 검색엔진 등의 다양한 소프트웨어를 통합, 운영하게 된다. 또한, 투입비용에 대한 효과적인 측면과 서비스 연속성 측면에서 서버의 공유 및 공유 스토로지에 대한 요구를 피할 수 없게 된다. 본 논문에서는 전체적인 서비스의 성능에 커다란 영향을 미치는 Back-end Tier의 검색서버를 통하여, IP 네트워크 기반의 스토리지 I/O에 대한 성능측정을 수행하였고, 그 성능실험의 결과에 따라 해당 애플리케이션에 최적의 스토리지 설계하고, 애플리케이션의 특성별로 최적의 스토리지를 할당하여 정보서비스의 전체적인 성능향상을 이루었다.

• The Journal of the Korea Contents Association
• /
• v.19 no.5
• /
• pp.187-193
• /
• 2019

Key-value storage engines are an essential component of growing demand in many computing environments, including social networks, online e-commerce, and cloud services. Recent key-value storage engines offer many features such as transaction, versioning, and replication. In a key-value storage engine, transaction processing provides atomicity through Write-Ahead-Logging (WAL), and a synchronous commit method for transaction processing flushes log data before the transaction completes. According to our observation, flushing log data to persistent storage is a performance bottleneck for key-value storage engines due to the significant overhead of fsync() calls despite the various optimizations of existing systems. In this article, we propose a group synchronization method to improve the performance of the key-value storage engine. We also design and implement a transaction scheduling method to perform other transactions while the system processes fsync() calls. The proposed method is an efficient way to reduce the number of frequent fsync() calls in the synchronous commit while supporting the same level of transaction provided by the existing system. We implement our scheme on the WiredTiger storage engine and our experimental results show that the proposed system improves the performance of key-value workloads over existing systems.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.9
• /
• pp.91-100
• /
• 2020

This paper presents an evaluation framework for prototyping multimedia streaming services including audio and video in a distributed and/or decentralized storage that can evaluate service quality and performance under various network conditions. The evaluation framework focuses on important indicators which measure and improve service quality by applying decentralized storage to multimedia streaming services that can mimic the scalability of the existing server-client software architecture and the issue of a single point of failure. The integrated framework not only measures performance indicators for evaluating the quality and performance of multimedia streaming on open source based multimedia content streaming services, but also adjusts network quality using network virtualization technology for comprehensive evaluations. The experimental results show that the integrated framework has low overhead in building and operating a decentralized storage with multimedia streaming services on a single host computer which validates the scalability of the developed framework.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.4
• /
• pp.15-20
• /
• 2016

Write amplification is a critical factor that limits the stable performance of flash-based storage systems. To reduce write amplification, this paper presents a new technique that cooperatively manages data in flash storage and nonvolatile memory (NVM). Our scheme basically considers NVM as the cache of flash storage, but allows the original data in flash storage to be invalidated if there is a cached copy in NVM, which can temporarily serve as the original data. This scheme eliminates the copy-out operation for a substantial number of cached data, thereby enhancing garbage collection efficiency. Experimental results show that the proposed scheme reduces the copy-out overhead of garbage collection by 51.4% and decreases the standard deviation of response time by 35.4% on average.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.2
• /
• pp.247-251
• /
• 2010

Using SCM in storage systems introduce new potentials for improving I/O performance and reliability. In this paper, we study the use of SCM as a buffer cache that guarantees transactional unit writes. Our proposed method can improve storage system reliability and performance at the same time and can recover the storage system immediately upon a system crash. The Proposed method is based on the LINUX JBD(Journaling Block Device), thus reliability is equivalent to JBD. In our experiments, the file system that adopts our method shows better I/O performance even while guaranteeing high reliability and shows fast file system recovery time (about 0.2 seconds).

• Convergence Security Journal
• /
• v.13 no.3
• /
• pp.85-90
• /
• 2013

To provide high-quality services, national public institutions and companies have provided information and materials over the internet network. However, a risk of malware infection between transmission and reception of data leads to exposure to various security threats. For this reason, national institutions have proceeded with projects for network separation since 2008, and data linkage has been made using network connection storage through network separation technologies, along with physical network separation. However, the network connection storage has caused waste of resources and problems with data management due to the presence of the same data in internal network storage and external network storage. In this regard, this study proposes a method to connect internal and external network data using NAS storage as a way to overcome the limitations of physical network separation, and attempts to verify the priority of items for the optimization of network separation by means of AHP techniques.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.6
• /
• pp.1539-1550
• /
• 2016

In modern society, the number of applications, which needs storage, is increased. Among them, the advance of FinTech increased the importance of storage encryption. FinTech storage, storing sensitive information, should be kept secure. Unless the storage is kept, many users will be damaged monetarily. To prevent this problem, we should encrypt the storage. A EncFS, which is one of the most popular storage encryption application, uses different IVs for each block to provide higher levels of security in the encryption algorithm. However, there is a vulnerability related to the usage of same IVs. In this paper, we propose a technique that decrypts the ciphertexts without knowing the secret key by using the vulnerability. Moreover, we show that the EncFS is not secure under IND-CPA model and propose a new scheme which is secure under IND-CPA model.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.11
• /
• pp.483-488
• /
• 2013

Users can access their data anywhere, at any time by using outsourced storage. But they cannot know how service provider manage the data. Even user cannot know when data damaged. To solve these problems, the outsourced storage auditing schemes has been proposed. Most proposed schemes are based on Homomorphic Verifiable Tags. But it has computational efficiency limitation because data used to exponent. In this paper, we propose a novel approach to outsourced storage auditing scheme using coefficient matrix. In the proposed scheme, data used to auditing by coefficient matrix form. Auditing procedures are proceed as solving the linear simultaneous equation. The auditor can audit easily by solving the equation using solution vector. The auditor can audit the n size data using sqrt(n) size data through out proposed scheme.

• The Journal of the Korea Contents Association
• /
• v.19 no.5
• /
• pp.417-425
• /
• 2019

Storage snapshot technology allows to preserve data at a specific point in time, and recover and access data at a desired point in time. It is an essential technology for storage protection application. Existing snapshot methods have some problems in that they dependent on storage hardware vendor, file system or virtual block device. In this paper, we propose a new snapshot method for solving the problems and creating snapshots on-line. The proposed snapshot method uses a method of extracting the log records of update operations at the virtual file system layer to enable the snapshot method to operate independently on file systems, virtual block devices, and storage hardwares. In addition, the proposed snapshot mehod creates and manages snapshots for directories and files without interruption to the storage service. Finally, through experiments we measure the snapshot creation time and the performance degradation caused by the snapshot.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.2
• /
• pp.77-82
• /
• 2022

Due to the recent advent of fast storage media, the memory management system needs to reconsider the configuring of a page unit. In this paper, we analyze the effect of the page size on memory performance as fast storage is adopted. Specifically, we analyze the TLB hit ratio and the page fault ratio as the workload and the page size are varied in desktop and smartphone environments. Our analysis shows that the influence of the page size depends on the system and workload conditions in desktop systems. However, in smartphone systems, the effect of the page size on memory performance is not large, and is not also sensitive to workloads. We expect that the analysis of this paper will be helpful in configuring the page size of given workloads under the system with fast storage media.