• Journal of Digital Convergence
• /
• v.20 no.5
• /
• pp.593-598
• /
• 2022

Recently, data production has seen explosive growth, and the storage systems to store these big data safely and quickly is evolving in various ways. A typical configuration of storage systems is the use of SSDs with fast data processing speed as a RAID group that can maintain reliable data. However, since NAND flash memory, which composes SSD, has the feature that deterioration if writes more than a certain number of times are repeated, can increase the likelihood of simultaneous failure on multiple SSDs in a RAID group. And this can result in serious reliability problems that data cannot be recovered. Thus, in order to solve this problem, we propose a method of throttling IOs so that each SSD within a RAID group leads to a different life-end. The technique proposed in this paper utilizes SMART to control the state of each SSD and the number of IOs allocated according to the data pattern used step by step. In addition, this method has the advantage of preventing large amounts of concurrency defects in RAID because it induces differential lifetime finishes of SSDs.

• The KIPS Transactions:PartD
• /
• v.9D no.2
• /
• pp.235-242
• /
• 2002

We propose an advanced transaction scheduling protocol for concurrency control of multilevel secure databases in wireless mobile network environment. Wireless communication is characterized by frequent spurious disconnections. So short-lived transaction must quickly access database without any delay by long-lived one. We adapted two-phase locking protocol, namely traditional syntax-oriented serializability notions, to multilevel secure databases in wireless mobile network environment. Altruistic locking, as an advanced protocol, has attempted to reduce delay effect associated with lock release moment by use of the idea of donation. An improved form of a1truism has also been deployed for extended a1truistic locking. This is in a way that scope of data to he early released is enlarged to include even data initially not intended to be donated. Our protocol is based on extended altruistic locking, but a new method, namely bi-directional donation locking for multilevel secure databases (MLBiDL), is additionally used in order to satisfy security requirements and concurrency. We showed the Simulation experiments that MLBiDL outperforms the other locking protocols in terms of the degree of throughput and average waiting time.

• Archives of design research
• /
• v.16 no.3
• /
• pp.183-190
• /
• 2003

It is a fundamental pre-requisite to thoroughly analyse and understand the things which are being designed in the process of industrial design. However, it is not always easy to acquire appropriate data to meet all the requirements to finally design a functionally superior products. This paper proposes an industrial design model with heightened reliability using the orthogonal array tables, which are fairly handy to apply when there are many design criteria to be considered at the onset stage Especially, in this research, the basic purpose of the orthogonal arrays that they try to compact the range of experiments and to improve the effectiveness of the experiment results is answered under average industrial design processes. At the same time, non-quantitative data of design factors are quantitative by the concurrency in design and their mutual actions are examined. This method can help industrial designers in narrowing their design possiblities by depicting more valid data, thus producing quality product designs by deriving optimal control factors.

• The Journal of the Korea Contents Association
• /
• v.7 no.7
• /
• pp.11-23
• /
• 2007

Applications involving moving objects require index structures to handle frequent updates of objects' locations efficiently. Several methods to index the current, the past and the future positions of moving objects have been proposed for the applications. Most of them are based on R-tree like index structures. Some researches have made efforts to improve update performance of R-trees that are actually focused on query performance. Even though the update performance is improved by researchers' efforts, the overhead and immaturity of concurrency control algorithms of R-trees makes us hesitate to choose them for moving objects. In this paper, we propose an update efficient indexing method that can be applicable for indexing the past, the current and the future locations. The proposed index is based on B+-Trees and Hilbert curve. We present an advanced Hilbert curve that adjusts automatically the order of Hilbert curve in subregions according to the data distribution and the number of data objects. Through empirical studies, we show that our strategy achieves higher response time and throughput.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.5
• /
• pp.468-476
• /
• 2000

Binding the object-oriented programming language C++ with a database management system provides a persistency to C++ programming objects so that objects can persist after program termina~,on. In such a binding system, we can manage a persistent object same as that we use a transient object and also use database management facilities such as transaction management and concurrency control. This paper presents a method providing the persistency to C++ programming objects in the binding system. We propose an improved interface based on C++ binding of ODMG-97 and present the design and implementation technique of it. The proposed interface provides a seamless interface for creating objects of the persistent capable class. We can create a persistent object without its class name as we do not give a class name when creating a transient object. Also, we guarantee the type compatibilty between the obiect created in database and the obiect created in main memory.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.147-154
• /
• 2021

Naval Combat Management System(a.k.a. CMS) is the core combat power of ROK Navy. CMS which has been localized since 1993 has been developed in various categories. However, in the characteristic of defense industry, CMS software has been rarely developed technically and structurally while the environment of computing system has been developed dramatically. A need for a new paradigm of CMS software development was raised. This paper suggests Naval Shield Component Platform(NSCP) as a standard interface architecture of CMS based on SOLID of OOP which is an advanced programming paradigm and introduce its functionality and feature. We expect NSCP's higher reusability, concurrency and maintainability in CMS software development. As a future work, we are going to apply NSCP to the next CMS software development project and evaluate quantitative, qualitative method.

• Journal of KIISE:Databases
• /
• v.29 no.1
• /
• pp.34-43
• /
• 2002

Since traditional static metadata set like Dublin Core has static metadata attributes about bibliography information, integration of metadata for various metadata, problems about standard and extension of metadata must be considered for applying it to new environment. Specially, as event-driven metadata write method included the notion of e-commerce come out for interoperability in digital libraries, traditional metadata management which cannot distinguish between different kinds of update operations to new extension of metadata set occurs unsuitable waiting of update operation. So, improvement is needed about it. In this paper, we show whether alleviative transaction consistency can be applied to digital library or not. Also it would divide newer metadata into static metadata attribute connected in read operation within user read-only transaction and dynamic metadata attribute in update operation within dynamic(e-commerce) update transactions. We propose newer metadata management algorithm considered in classfication of metadata attributes and dynamic update transaction. Using two version for minimal maintenance cost and ARU(Appended Refresh Unit) for dynamic update transaction, to minimize conflict between read and write operations shows fast response time and high recency ratio. As a result of the performance evaluation, we show our algorithm is proved to be better than other algorithms in newer metadata environments.