• KIISE Transactions on Computing Practices
• /
• v.22 no.2
• /
• pp.95-100
• /
• 2016

In the current era of big data and cloud computing, the amount of data utilized is increasing, and various systems to process this big data rapidly are being developed. A distributed file system is often used to store the data, and glusterFS is one of popular distributed file systems. As computer technology has advanced, NAND flash SSDs (Solid State Drives), which are high performance storage devices, have become cheaper. For this reason, datacenter operators attempt to use SSDs in their systems. They also try to install glusterFS on SSDs. However, since the glusterFS is designed to use HDDs (Hard Disk Drives), when SSDs are used instead of HDDs, the performance is degraded due to structural problems. The problems include the use of I/O-cache, Read-ahead, and Write-behind Translators. By removing these features that do not fit SSDs which are advantageous for random I/O, we have achieved performance improvements, by up to 255% in the case of 4KB random reads, and by up to 50% in the case of 64KB random reads.

• Journal of KIISE
• /
• v.43 no.4
• /
• pp.411-418
• /
• 2016

In recent times, NAND flash memory has become widely used as secondary storage for computing devices. Accordingly, to take advantage of NAND flash memory, new file systems have been actively studied and proposed. The performance of these file systems is generally measured with benchmark tools. However, since benchmark tools are executed by software simulation methods, many researchers get non-uniform benchmark results depending on the system environments. In this paper, we use Filebench, one of the most popular and representative benchmark tools, to analyze benchmark results and study the reasons why the benchmark result variations occur. Our experimental results show the differences in benchmark results depending on the system environments. In addition, this study substantiates the fact that system performance is affected mainly by background I/O requests and fsync operations.

• The Journal of Engineering Research
• /
• v.7 no.1
• /
• pp.71-78
• /
• 2005

The nowaday Internet reached in phase that collect ideas of netizens through the Internet now regularly according as offer netizens great many information. In these circumstance, web survey system knows off-line just in on-line the value more do can. However, because participations of many people are important, web survey must be subject that can attract interest of people, and more visual function sees that can lead high hailing degree in active interest and survey more than netizen by offering new interface that provoke interest because is added than uninteresting screen. Therefore, I wish to enhance value still more by doing as can do survey interestingly using web survey system that use flash.

• Transactions on Electrical and Electronic Materials
• /
• v.17 no.1
• /
• pp.25-28
• /
• 2016

We investigate the effect of interface states on the endurance of a charge trap flash (CTF) NOR array using charge pumping methods. The endurance test was completed from one cell selected randomly from 128 bit cells, where the memory window value after 10² program/erase (P/E) cycles decreased slightly from 2.2 V to 1.7 V. However, the memory window closure abruptly accelerated after 10³ P/E cycles or more (i.e. 0.97 V or 0.7 V) due to a degraded programming speed. On the other hand, the interface trap density (Nit) gradually increased from 3.13×10¹¹ cm⁻² for the initial state to 4×10¹² cm⁻² for 10² P/E cycles. Over 10³ P/E cycles, the Nit increased dramatically from 5.51×10¹² cm⁻² for 10³ P/E cycles to 5.79×10¹² cm⁻² for 10⁴ P/E cycles due to tunnel oxide damages. These results show good correlation between the interface traps and endurance degradation of CTF devices in actual flash cell arrays.

• Journal of Korean Society of Water and Wastewater
• /
• v.19 no.3
• /
• pp.370-377
• /
• 2005

Dispersion of coagulant should be completed in a fraction of a second before the metal hydroxide precipitate has form. For the reason so-called pump diffusion flash mixing (PDFM) have been proposed, and PDFM is one of reasonable methods to quickly disperse the hydrolyzing metal salts. In this study, therefore, we attempt to understand the difference of removal characteristics of natural organic matter (NOM) between pump diffusion flash mixing (PDFM) and conventional rapid mixing (CRM) for coagulation in a water treatment system, and to enhance the removal of NOM through the improved mixing process. DOC and turbidity removal by PDFM higher than those by CRM, while SUVA value of water treated by PDFM was high as compared with that by CRM. Hydrophilic NOM was more effectively removed by PDFM than CRM, since charge neutralization effect increased by quick dispersion of coagulant. The DBP formation potentials due to NOM was effectively reduced by the improved mixing (i.e., PDFM) for coagulation and could be controlled through decrease in concentration of precursor rather than reduction of activity with disinfectant.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.11
• /
• pp.2082-2089
• /
• 1992

An experimental study on forging an axi-symmetric dome type of AISI4130 was carried out using modeling material. In order to verify the validity of the experimental data, a similarity study between plasticine and AISI4130 has been made. Friction conditions were characterized by ring test for the various lubricants. For the closed-die forging experiments of an axi-symmetric dome type of AISI4130 using the plasticine, various cylindrical billets with different aspect ratios were forged and different flash width to thickness(W/T) ratios were used in order to determine the optimum forging conditions. As W/T ratios decrease forging loads decrease while excess volumes increase. It was found out that the experimental results reproduce the similiar results available in the literature. As a result of these experiments, it was construed physical modeling is an excellent tool for forging process simulation at a practical level.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.12
• /
• pp.419-428
• /
• 2022

Since in-place updates for pages are not allowed in flash memory, all new page writes should be written in an out-of-place manner. The old overwritten pages are invalidated. Such invalidated pages eventually trigger the costly garbage collection process. Since the garbage collection causes numerous read and write operations, it is one of the flash memory's major performance issues. In 2R, it modified the garbage collection algorithm, which applies the I/O characteristics of the On-Line Transaction Process workload to improve the Write Amplification Factor. However, this algorithm has a region pollution problem. Therefore, in this paper, we developed 2R++ that additionally separates pages with long access intervals to solve the region pollution problem. 2R++ introduces an extra bit per block to separate warm pages based on a second chance mechanism. Prevents warm pages from being misidentified as cold pages to solve region pollution problem. We conducted the experiments on TPC-C and Linkbench to make the performance comparison. The experiment showed that 2R++ achieved a Write Amplification Factor improvement of 57.8% and 13.8% compared to 2R, respectively.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.05a
• /
• pp.1063-1064
• /
• 2010

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06a
• /
• pp.333-334
• /
• 2008

• Journal of KIISE
• /
• v.44 no.5
• /
• pp.469-475
• /
• 2017

The emerging NVMe-based multi-queue SSDs provides a high bandwidth by parallel I/O, i.e., each core performs I/O through its dedicated queue in parallel with other cores. To provide a bandwidth share for each application with I/O, a fair-share scheduler that provides a bandwidth share to each core is required. In this study, we proposed a multi-core scalable fair-queuing algorithm for multi-queue SSDs. The algorithm adopts randomization to minimize the inter-core synchronization overheads and provides a weight-proportional bandwidth share to each core. The results of our experiments indicated that the proposed algorithm gives accurate bandwidth partitioning and outperforms the existing FlashFQ scheduler, regardless of the number of cores for a Linux kernel with block-mq.