• 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 the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.33 no.6
• /
• pp.454-459
• /
• 2020

In this study, a TiO₂/TiO_2-x-based resistance variable memory was fabricated using a DC/RF magnetron sputtering system and ALD. In order to analyze the effect of oxygen plasma treatment on the performance of resistance random access memory (ReRAM), the TiO₂/TiO_2-x-based ReRAM was evaluated by applying RF power to the TiO_2-x oxygen-holding layer at 30, 60, 90, 120, and 150 W, respectively. The ReRAM was fabricated, and the electrical and surface area performances were compared and analyzed. In the case of ReRAM without oxygen plasma treatment, the I-V curve had a hysteresis curve shape, but the width was very small, with a relatively high surface roughness of the oxygen-retaining layer. However, in the case of oxygen plasma treatment, the HRS/LRS ratio for the I-V curve improved as the applied RF power increased; stable improvement was also noted in the surface roughness of the oxygen-retaining layer. It was confirmed that the low voltage drive was not smooth due to charge trapping in the oxygen diffusion barrier layer owing to the high intensity ReRAM applied with an RF power of approximately 150 W.

• Korean Journal of Materials Research
• /
• v.14 no.9
• /
• pp.649-654
• /
• 2004

In this paper, the fracture toughness and mechanisms of failure in a random SiC-whisker/$Al_{2}O_3$ ceramic composite were investigated using in situ observations during mode I(opening) loading. $SiC_{w}/Al_{2}O_3$ composite was obtained by hot press sintering of $Al_{2}O_3$ powder and SiC whisker as the matrix and reinforcement, respectively. The whisker and powder were mixed using a turbo mill. The composite was produced at SiC whisker volume fraction of $0.3\%$. Compared with monolithic $Al_{2}O_3$, fracture toughness enhancement was observed in $SiC_{w}/Al_{2}O_3$ composite. This improved fracture toughness was attributed to SiC whisker bridging and crack deflection. $SiC_{w}/Al_{2}O_3$ composite exhibited typically brittle fracture behavior, but a fracture process zone was observed in this composite. This means that the load versus load-line displacement curve of $SiC_{w}/Al_{2}O_3$ composite from a fracture test may involve a small non-linear region near the peak load.

• Journal of Digital Convergence
• /
• v.12 no.1
• /
• pp.379-388
• /
• 2014

We noted that substituting hard disk with high-performance storage device on SAN did not immediately result in getting high performance. Investigating the reason behind this leaded us to propose optimization schemes for high-performance storage system. We first got rid of the latency in the I/O process which is unsuitable for the high-performance storage device, added parallelism on the storage server, and applied temporal merge to Superhigh speed network protocol for improving the performance with small random I/O. The proposed scheme was implemented on the SAN with high-performance storage device and we verified that there were about 30% reduction on the I/O delay latency and 200% improvement on the storage bandwidth.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.2
• /
• pp.201-205
• /
• 2008

The flash memory is widely used as a main storage of embedded devices. It is adopted as a storage of database as growing the capacity of the flash memory. We run TPC-C benchmark on various FTL algorithms. But, the database shows poor performance on flash memory because the characteristic of I/O requests is full random. In this paper, we show the performance of all existing FTL algorithms is very poor. Especially, the FTL algorithm known as good at small mobile equipment shows worst performance. In addition, the chip-inter leaving which is a technique to improve the performance of the flash memory doesn't work well. In this paper, we inform you the reason that we need a new FTL algorithm and the direction for the database in the future.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.8C
• /
• pp.670-679
• /
• 2012

The recent dissemination of cloud computing makes the amount of data storage to be increased and the cost of storing the data grow rapidly. Accordingly, data and service requests from users also increases the load on the cloud storage. There have been many works that tries to provide low-cost and high-performance schemes on distributed file systems. However, most of them have some weaknesses on performing parallel and random data accesses as well as data accesses of frequent small workloads. Recently, improving the performance of distributed file system based on caching technology is getting much attention. In this paper, we propose a CHPC(Cloud storage High-Performance Caching) framework, providing parallel caching, distributed caching, and proxy caching in distributed file systems. This study compares the proposed framework with existing cloud systems in regard to the reduction of the server's disk I/O, prevention of the server-side bottleneck, deduplication of the page caches in each client, and improvement of overall IOPS. As a results, we show some optimization possibilities on the cloud storage systems based on some evaluations and comparisons with other conventional methods.