• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.4
• /
• pp.368-380
• /
• 2005

File systems for flash memory that is widely used as a storage device for mobile devices should provide not only high-performance data reads and writes but also a guarantee on the data integrity even on a power failure. In this paper, we explain the design and implementation of a file system for flash memory that considers flash memory's physical characteristics and the data layout in the file system to give an optimized write performance. This file system guarantees the reliability against various system failures including a power failure by using the transaction concept in write processing. In addition, the file system minimizes the memory usage by using a simple static mapping. In the paper, we also describe the implementation of the file system and compare its performance with other existing flash memory ille systems.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.2
• /
• pp.98-113
• /
• 2009

Conventional file systems for harddisk drive cannot be applied to NAND flash memory, because the physical characteristics of NAND flash memory differs from those of harddisk drive. To address this problem, various file systems with better reliability and efficiency have also been developed recently. However, those file systems have inherent overheads for updating the file's metadata pages, because those file systems save file's meta-data and data separately. Furthermore, those file systems have a critical reliability problem: file systems fail when either a page in meta-data of a file system or a file itself fails. In this paper, we propose a self-description page technique and In Memory Core File System technique to address these efficiency and reliability problems, and develop SDFS(Self-Description File System) newly. SDFS can be safely recovered, although some pages fail, and improves write and read performance by 36% and 15%, respectively, and reduces mounting time by 1/20 compared with YAFFS2.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.6
• /
• pp.594-598
• /
• 2008

In this paper we propose a fast booting technique based on Improved snapshot boot in embedded Linux, widely adopted in personal devices such as PDA and mobile phones. The existing Snapshot boot technique tries to create a snapshot image at the time of suspend, and later load the entire snapshot image into the system memory at the predefined location with the help of a bootloader at the time of resume. Since a bootloader has to copy the entire snapshot image into the predefined memory to resume the previous suspended computing state, a little bit long time is required to resume. Improved snapshot boot does not create a snapshot image consisting of whole memory pages at the time of suspend, thus resulting in smaller snapshot image than the existing snapshot boot. The remaining pages are in the swap area. The resulting smaller sized snapshot image enables much faster booting latency. Through the experiment, we can see the booting latency is reduced almost 30% with suspend image of 2982 pages. This result depends on the amount of swap-out pages.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.12
• /
• pp.1199-1203
• /
• 2010

Data transfer between host system and storage device is based on the data unit called sector, which can be varied depending on computer systems. If NAND flash memory is used as a storage device, the variant sector size can affect storage system performance since its operation is much related to sector size and page size. In this paper, we propose an efficient FTL mapping management scheme to support multiple sector size within one NAND flash memory based storage device, and analyze the performance effect and management overhead. According to the proposed scheme, the management overhead of proposed FTL management is lower than conventional scheme when various sector sizes are configured in computer systems, while performance is less degraded in comparison with single sector size support system.

• Journal of Internet Computing and Services
• /
• v.6 no.1
• /
• pp.73-84
• /
• 2005

ALTIBASE/sup TM/ is the relational main-memory DBMS in which a main memory is primarily used as the main storage device. We present the database replication strategies and techniques of the ALTIBASETM system in order to meet the requirement of high availability and efficient transaction processing. Our process architecture for replication management and its communication model are proposed, and database replication protocols are also described. We show the experimental result of transaction processing rate with various DBMS parameters and overall performance of database replication system as compared to standalone system.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.9
• /
• pp.685-689
• /
• 2009

Flash memory has several features such as low~power consumption and fast access so that there has been various research on using flash memory as new storage. Especially the Solid State Disk which is composed of flash memory chips has recently replaced the hard disk. At present, SSD adopts the multi-channel and multi-way architecture to exploit advantages of parallel access. In this architecture, data are written on SSD in a unit of a superblock which is composed of multiple blocks in which some blocks are put together. This paper proposes two schemes of selecting, segmenting and re-composing victim superblocks to optimize concurrent processing when a buffer flush occurs. The experimental results show that 35% of superblock- based write operations is reduced by selecting victims and additional 9% by composition of superblock.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.4
• /
• pp.254-263
• /
• 2006

Ontology-based context models are widely used in ubiquitous computing environment because they have advantages in the acquisition of conceptual context through inferencing, context sharing, and context reusing. Among the benefits, inferencing enables context-aware applications to use conceptual contexts which cannot be acquired by sensors. However, inferencing causes processing delay and thus becomes the major obstacle to the implementation of context-aware applications. The delay becomes longer as the amount of contexts increases. In this paper, we propose a context pre-fetching method to reduce the size of contexts to be processed in a working memory in attempt to speed up inferencing. For this, we extend the query-tree method to identify contexts relevant to the queries of a context-aware application. Maintaining the pre-fetched contexts optimal in a working memory, the processing delay of inference reduces without the loss of the benefits of ontology-based context model. We apply the proposed scheme to our ubiquitous computing middleware, Active Surroundings, and demonstrate the performance enhancement by experiments.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.541-542
• /
• 2021

클라우드 컴퓨팅 환경에서는 고성능 컴퓨팅을 지원하기 위해 사용자에게 GPU(Graphic Processing Unit)가 할당된 가상머신을 제공하여 사용자가 고성능 응용을 실행할 수 있도록 지원한다. 일반적인 컴퓨팅 환경에서 한 명의 사용자가 GPU를 독점해서 사용하기 때문에 자원 경쟁으로 인한 문제가 상대적으로 적게 발생하지만 독립적인 여러 사용자가 컴퓨팅 자원을 공유하는 클라우드 환경에서는 자원 경쟁으로 인해 서로 성능 영향을 미치는 문제를 발생시킨다. 본 논문에서는 여러 개의 가상머신이 단일 GPU를 공유하는 RPC(Remote Procedure Call) 기반 GPU 가상화 환경에서 다수의 가상머신이 GPGPU(General Purpose computing on Graphics Processing Units) 작업을 수행할 때 GPU 메모리 입력 경쟁으로 인해 발생하는 커널 함수의 실행 지연 문제를 분석한다.

• Journal of KIISE
• /
• v.44 no.10
• /
• pp.1005-1018
• /
• 2017

Recently, virtualization has become an essential component of cloud computing due to its various strengths, including maximizing server resource utilization, easy-to-maintain software, and enhanced data protection. However, since virtualization allows sharing physical resources among the VMs, the system performance can be deteriorated due to device contentions. In this paper, we first investigate the I/O overhead based on the number of VMs on the same server platform and analyze the block I/O process of the KVM hypervisor. We also propose an in-memory block cache mechanism, called QBic, to overcome I/O virtualization latency. QBic is capable of monitoring the block I/O process of the hypervisor and stores the data with a high access frequency in the cache. As a result, QBic provides a fast response for VMs and reduces the I/O contention to physical devices. Finally, we present a performance measurement of QBic to verify its effectiveness.

• Journal of Computing Science and Engineering
• /
• v.5 no.2
• /
• pp.111-120
• /
• 2011

Data mining algorithms should exploit new hardware technologies to accelerate computations. Such goal is difficult to achieve in database management system (DBMS) due to its complex internal subsystems and because data mining numeric computations of large data sets are difficult to optimize. This paper explores taking advantage of existing multithreaded capabilities of multicore CPUs as well as caching in RAM memory to efficiently compute summaries of a large data set, a fundamental data mining problem. We introduce parallel algorithms working on multiple threads, which overcome the row aggregation processing bottleneck of accessing secondary storage, while maintaining linear time complexity with respect to data set size. Our proposal is based on a combination of table scans and parallel multithreaded processing among multiple cores in the CPU. We introduce several database-style and hardware-level optimizations: caching row blocks of the input table, managing available RAM memory, interleaving I/O and CPU processing, as well as tuning the number of working threads. We experimentally benchmark our algorithms with large data sets on a DBMS running on a computer with a multicore CPU. We show that our algorithms outperform existing DBMS mechanisms in computing aggregations of multidimensional data summaries, especially as dimensionality grows. Furthermore, we show that local memory allocation (RAM block size) does not have a significant impact when the thread management algorithm distributes the workload among a fixed number of threads. Our proposal is unique in the sense that we do not modify or require access to the DBMS source code, but instead, we extend the DBMS with analytic functionality by developing User-Defined Functions.