• The KIPS Transactions:PartD
• /
• v.17D no.6
• /
• pp.395-404
• /
• 2010

Solid-State Disks (SSDs) have been currently considered as a promising candidate to replace hard disks, due to their significantly short access time, low power consumption, and shock resistance. SSDs, however, have drawbacks such that their write throughput and life span are decreased by random-writes, nearly regardless of SSDs controller designs. Previous studies have mostly focused on better designs of SSDs controller and reducing the number of write operations to SSDs. We suggest another method that reallocates data pages that tend to be simultaneously written to contiguous blocks. Our method gathers write operations during a period of time and generates write traces. After transforming each trace to a set of transactions, our method mines frequent itemsets from the transactions and reallocates the pages of the frequent itemsets. In addition, we introduce an algorithm that reallocates the pages of the frequent itemsets with moderate time complexity. Experiments using TPC-C workload demonstrated that our method successfully reduce 6% of total logical block access.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.7
• /
• pp.159-166
• /
• 2019

Unlike modern Solid-State Drives with DRAM, DRAM-less SSDs do not have DRAM because they are cheap and consume less power. Obviously, they have performance degradation problem due to lack of DRAM in the controller and this problem can be alleviated by utilizing host memory buffer(HMB) feature of NVMe, which allows SSDs to utilize the DRAM of host. In this paper, we show that commercial DRAM-less SSDs surely exhibit lower I/O performance than other SSDs with DRAM, but they can be improved by utilizing the HMB feature. Through various experiments and analysis, we also show that DRAM-less SSDs mainly exploit the DRAM of host as mapping table cache rather than read cache or write buffer to improve I/O performance.

• 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.

• The KIPS Transactions:PartD
• /
• v.17D no.6
• /
• pp.383-394
• /
• 2010

Flash memory based SSDs are currently being considered as a promising candidate for replacing hard disks due to several superior features such as shorter access time, lower power consumption and better shock resistance. However, SSDs have different characteristics from hard disk such as difference of unit and time for read, write and erase operation and impossibility for over-writing. Because of these reasons, SSDs have disadvantages on hard disk based systems, so FTL(Flash Translation Layer) is designed to increase SSDs' efficiency. In this paper, we propose an advanced logical address mapping method for increasing SSDs' performance, which is named HAMM(Hybrid Address Mapping Method). HAMM addresses drawbacks of previous block-mapping method and super-block-mapping method and takes advantages of them. We experimented our method on our own SSDs simulator. In the experiments, we confirmed that HAMM uses storage area more efficiently than super-block-mapping method, given the same buffer size. In addition, HAMM used smaller memory than block-mapping method to construct mapping table, demonstrating almost same performance.

• Economic and Environmental Geology
• /
• v.52 no.5
• /
• pp.471-484
• /
• 2019

Liquefaction occurs by a temporal loss of sediment strength as a consequence of increased pore water pressure during the re-arrangement of unconsolidated, granular sediments. Liquefaction is dependent on the physical properties of the sediments and cause surface cracks, landslide, and the formation of soft-sediment deformation structures(SSDS). SSDS is formed by the combined action of the driving force and deformation mechanism(liquefaction, thixotropy, and fluidization) that is triggered by endogenic or exogenic triggers. So research on the SSDS can unravel syndepositional geological events. If detailed sedimentologic analysis together with surrounding geological context suggest SSDS formed by earthquakes, the SSDS provide a clue to unravel syndepositional tectonic activities and detailed paleoseismological information(> Mw 5) including earthquakes that leave no surface expression.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.15 no.5
• /
• pp.511-518
• /
• 2015

As the cost-per-byte of SSDs dramatically decreases, the introduction of SSDs to Hadoop becomes an attractive choice for high performance data processing. In this paper the cost-per-performance of SSD-based Hadoop cluster (SSD-Hadoop) and HDD-based Hadoop cluster (HDD-Hadoop) are evaluated. For this, we propose a MapReduce performance model using queuing network to simulate the execution time of MapReduce job with varying cluster size. To achieve an accurate model, the execution time distribution of MapReduce job is carefully profiled. The developed model can precisely predict the execution time of MapReduce jobs with less than 7% difference for most cases. It is also found that SSD-Hadoop is 20% more cost efficient than HDD-Hadoop because SSD-Hadoop needs a smaller number of nodes than HDD-Hadoop to achieve a comparable performance, according to the results of simulation with varying the number of cluster nodes.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.2
• /
• pp.71-78
• /
• 2019

As the capacity of SSDs rapidly increases, the amount of DRAM to keep a mapping table size in SSDs becomes very huge. To address a Demand-based FTL (DFTL) scheme that caches part of mapping entries in DRAM is considered to be a feasible alternative. However, owing to its unpredictable behaviors, DFTL fails to provide consistent I/O response times. In this paper, we a) analyze a root cause that results in fluctuation on read latency and b) propose a new demand-based FTL scheme that ensures guaranteed read response time with low write amplification. By preventing mapping evictions while serving reads, the proposed technique guarantees every host read requests to be done in 2 NAND read operations. Moreover, only with 25% of a cache ratio, the proposed scheme improves random write performance and random mixed performance by 1.65x and 1.15x, respectively, over the traditional DFTL.

• Journal of KIISE
• /
• v.44 no.2
• /
• pp.208-212
• /
• 2017

Flash SSDs have many advantages over the existing hard disks such as energy efficiency, shock resistance, and high I/O throughput. For these reasons, in combination with the emergence of innovative technologies such as 3D-NAND and V-NAND for cheaper cost-per-byte, flash SSDs have been rapidly replacing hard disks in many areas. However, the existing database engines, which have been developed mainly assuming hard disks as the storage, could not fully exploit the characteristics of flash SSDs (e.g. internal parallelism). In this paper, in order to utilize the internal parallelism intrinsic to modern flash SSDs for faster query processing, we implemented a prefetching method using asynchronous input/output as a new functionality for secondary index scans in MySQL InnoDB engine. Compared to the original InnoDB engine, the proposed prefetching-based scan scheme shows three-fold higher performance in the case of 16KB-page sizes, and about 4.2-fold higher performance in the case of 4KB-page sizes.

• The Journal of the Korea Contents Association
• /
• v.17 no.7
• /
• pp.93-100
• /
• 2017

Recently, demands for high-performance flash-based storage devices (i.e., flash SSD) have rapidly grown in social network services, cloud computing, super-computing, and enterprise storage systems. The industry and academic communities made the NVMe specification for high-performance storage devices, and NVMe-based flash SSDs can be now obtained in the market. In this article, we evaluate performance of NoSQL databases that social network services and cloud computing services heavily adopt by using NVMe-based flash SSDs. To this end, we use NVMe SSD that Samsung Electronics recently developed, and the SSD used in this study has performance up to 3.5GB/s for sequential read/write operations. We use WiredTiger for NoSQL databases, and it is a default storage engine for MongoDB. Our experimental results show that log processing in NoSQL databases is a major overhead when high-performance NVMe-based flash SSDs are used. Furthermore, we optimize components of log processing and optimized WiredTiger show up to 15 times better performance than original WiredTiger.

• Smart Structures and Systems
• /
• v.30 no.5
• /
• pp.479-500
• /
• 2022

In this study, a new type of isolation bearing is proposed by combining S-shaped steel plate dampers (SSDs) with a spherical steel bearing, and the seismic control effect of a five-span standard high-speed railway bridge is investigated. The advantages of the proposed S-shaped steel damping friction bearing (SSDFB) are that it cannot only lengthen the structural periods, dissipate the seismic energy, but also prevent bridge unseating due to the restraint effectiveness of SSDs in the large relative displacements between the girders and piers. This study first presents a detailed description and working principle of the SSDFB. Then, mechanical modeling of the SSDFB was derived to fundamentally define its cyclic behavior and obtain key mechanical parameters. The numerical model of the SSDFB's critical component SSD was verified by comparing it with the experimental results. After that, parameter studies of the dimensions and number of SSDs, the friction coefficient, and the gap length of the SSDFBs were conducted. Finally, the longitudinal seismic responses of the bridge with SSDFBs were compared with the bridge with spherical bearing and spherical bearing with strengthened shear keys. The results showed that the SSDFB can not only significantly mitigate the shear force responses and residual displacement in bridge substructures but also can effectively reduce girder displacement and prevent bridge unseating, at a cost of inelastic deformation of the SSDs, which is easy to replace. In conclusion, the SSDFB is expected to be a cost-effective option with both multi-stage energy dissipation and restraint capacity, making it particularly suitable for seismic isolation application to high-speed railway bridges.