• The Transactions of the Korea Information Processing Society
• /
• v.3 no.1
• /
• pp.87-94
• /
• 1996

In order to design an extended relational database management system supporting multilevel security, the standard relational data model is extended and new relational integrity constrains are proposed for the model. The extended relational model and proposed multilevel integrity constraniants maintain database in consistent state and produce a basis that can eliminat eambiguity of entity and relation ship representations bypoly instantiation. The proposed up dates emantics canincreases the efficiency of up date operations by supporting multilevel entry and up dates. The semantics also provides a basis for the implementation of decomposition of extended relations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.7
• /
• pp.1561-1568
• /
• 2005

In client-solver environments, clients can execute applications without contact a server by storing data in its cache. To guarantee correction of applications, we need a cache consistency algorithm. Many cache consistency algorithms have been proposed, these were categorized by detection-based and avoidance-based algorithms. In this paper, we propose a new detection-based cache consistency algorithm that can reduce a burden of a client that has poor resource compare with a server Our method controls concurrency of transactions using timestamp ordering and updates a cache dynamically. In addition to we show that our method maintains a one-copy serializability.

• Journal of KIISE:Databases
• /
• v.32 no.6
• /
• pp.628-640
• /
• 2005

The XML is well accepted in several different Web application areas. As soon as many users and applications work concurrently on the same collection of XML documents, isolating accesses and modifications of different transactions becomes an important issue. When an XML document correctly corresponds to the rules laid out in a DTD or XML schema, it is also said to be valid. The valid XML document's validity should be guaranteed after the document is updated. The validation method mentioned above, however, results in lower degree of concurrency. For getting higher degree of concurrency and minimizing the range of the XML document validity, a new validation method based on a specific locking method is required. In this paper we propose the sequence group validation method for minimizing the range of the XML document validity. We also propose the boundary locking method for isolating accesses and modifications of different transactions while supporting the valid XML document's validity. Finally, the results of some experiments show the validation and locking methods increase the degree of transaction concurrency.

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.140-149
• /
• 2004

Data broadcast has received much attention as a very efficient method for disseminating data items in mobile environment with large number of mobile clients. In this approach, a database server periodically and continuously broadcasts data items through wireless channels and clients perform read-only transactions by accessing necessary data items from the air. While broadcasting, the server must also process update transactions on the database, which raises an obstacle for client's accessing consistent data. In this research, we propose a new algorithm SCDSC(Serialization Checking with DirtySet on Commit) which is an alternative for solving the concurrency control problem efficiently. The SCDSC is a kind of optimistic concurrency control in that a client checks the consistency of data using a DirtySet as a part of data broadcast when it commits its transaction. In each broadcast cycle, the server updates and disseminates the DirtySet with newly changed data items for last few cycles in the sliding window approach. We perform an analysis and a simulation study to evaluate the performance of our SCDSC algorithm in terms of data consistency and data currency.

• Journal of KIISE:Databases
• /
• v.29 no.1
• /
• pp.1-14
• /
• 2002

Association rule discovery is a method of mining for the associated item set on large databases based on support and confidence threshold. The discovered association rules can be applied to the marketing pattern analysis in E-commerce, large shopping mall and so on. The association rule discovery makes multiple scan over the database storing large transaction data, thus, the algorithm requiring very high overhead might not be useful in real-time association rule discovery in dynamic environment. Therefore this paper proposes an active candidate set management model based on trigger and incremental update mechanism to overcome non-realtime limitation of association rule discovery. In order to implement the proposed model, we not only describe an implementation model for incremental updating operation, but also evaluate the performance characteristics of this model through the experiment.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.4
• /
• pp.85-91
• /
• 2019

In this paper, we propose a new logging and recovery scheme that is suited for the high-performance transaction processing system base on flash memory storage. The proposed scheme is designed by considering flash's I/O characteristic of asymmetric costs between page update/read operations. That is, we substitute the costly update operation with writing and real-time usage of snapshot log, which is for the page-level physical redo. From this, we can avoid costly rewriting of a dirty page when it is evicted form a buffering pool. while supporting efficient revery procedure. The proposed scheme would be not lucrative in the case of HDD-based system. However, the proposed scheme offers the performance advance sush as a reduced number of updates and the fast system recovery time, in the case of flash storage such as SSD (solid state drive). Because the proposed scheme can easily be applied to existing systems by saving our snapshot records and ordinary log records together, our scheme can be used for improving the performance of upcoming SSD-based database systems through a tiny modification to existing REDO algorithms.

• The Transactions of the Korea Information Processing Society
• /
• v.1 no.2
• /
• pp.154-162
• /
• 1994

In this paper, a new locking protocol which can increase the concurrency in an Object Oriented Database Systems is proposed. The proposed locking protocol uses the Root object identifier to identify the root of a component object in a composite object. The transaction that requests a lock on the component object checks whether the Rid of the component object has been locked by another transaction. Even though the class of the component object is locked by one transaction, other transactions can concurrently access other instances in the class, if each Rid of them is not locked by any other transactions. This can increase the degree of concurrency.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.3
• /
• pp.659-669
• /
• 1997

Security as weel as consistency of data is very important issue in databaed security.This the serializability of transactions must be maintained and particularly covert channel not caesed between a high-level transaction and a low- level one.In this paper, we propose a secure transaction management algorithm using synamic version control] method that can slove disk space overhead to maintain multiversion and the problem that reansactions read too old versions when two versions are maintained.Disk space overhead can be sloved by properly cerating versions and synamically maintaining the number of versions and the problem for reading too old version can be solved by having transactions read versions as recent as possible.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.1043-1046
• /
• 2006

Cached data management of clients is required to guarantee the correctness of client's applications. There are two categories of cache consistency algorithms : detection-based and avoidance-based cache consistency algorithm. Detection?.based schemes allow stale data access and then check the validity of any cached data before they ran be allowed to commit. In contrast, under avoidance-based algorithms, transactions never have the opportunity to access stale data. In this paper, we propose a new avoidance-based cache consistency algorithm make use of version. The proposed method maintains the two versions at clients and servers, so it has no callback message and it can be reduced abort ratio of transactions compare with the single-versioned algorithms. In addition to, the proposed method can be decreased cache miss using by mixed invalidation and propagation for remote update action.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.101-104
• /
• 2006

다중 레벨 공간 DBMS 는 디스크 기반 DBMS 와 메인 메모리 저장관리자를 통합한 시스템이다. 다중 레벨 공간 DBMS 는 디스크 기반 DBMS 를 사용하여 대용량 데이터를 저장하고 자주 이용되는 데이터만 스냅샷으로 메인 메모리에 상주 시켜 빠르게 트랜잭션을 처리한다. 그러나 시스템 운영상에서 스냅샷의 중복 생성 및 부적절한 갱신 주기의 문제가 발생하여 시스템의 성능을 저하시킨다. 그래서 본 논문에서는 스냅샷 데이터 관리를 위한 관리도구를 제안한다. 본 관리도구는 스냅샷의 생성 및 삭제, 동기화 등의 조작 기능을 제공하고, 메타데이터를 관리하여 통계정보 및 스냅샷 상세정보를 제공하여 최적화를 위한 판단기준으로 사용 할 수 있도록 한다. 본 도구를 이용하면 관리자가 시스템이 발견하지 못한 스냅샷의 중복된 영역 및 최적인 갱신 주기를 판단 하고 스냅샷을 조작해 시스템의 성능을 보다 최적 할 수 있다.