Using data clustering to improve cleaning performance for flash memory

Chiang, Mei‐Ling; Lee, Paul C. H.; Chang, Ruei-Chuan

doi:10.1002/(sici)1097-024x(199903)29:3<267::aid-spe233>3.0.co;2-t

Cited by 122 publications

(79 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Chiang et al proposed Dynamic Age Clustering (DAC) [5], which, similarly to eNVy, shares the overall design of partitioning to space in regions but, unlike eNVy, uses different page migration policies between regions. Pages are promoted to a colder region when garbage collected, and to a hotter region when updated and if their age (measured in seconds) is smaller than a workload dependent threshold.…”

Section: State-of-art Data Placement Algorithmsmentioning

confidence: 99%

“…Previous proposals rely on variable parameters that are workload specific. Examples include the number of regions [12,5,26], the time threshold to migrate data from one region to the next [5], the empirical probabilistic model of [12].…”

Section: Data Placement Challengementioning

confidence: 99%

“…FTLs are usually categorized based on the unit size of the logical to physical mapping: page level mapping FTLs [26,5,10,12,18], hybrid page-block level FTLs [15,16]), or variable size mapping FTLs [17,6].…”

Section: Applicabilitymentioning

confidence: 99%

See 2 more Smart Citations

Improving flash write performance by using update frequency

Stoica

Ailamaki

2013

Proc. VLDB Endow.

View full text Add to dashboard Cite

Solid-state drives (SSDs) are quickly becoming the default storage medium as the cost of NAND flash memory continues to drop. However, flash memory introduces new challenges, as data cannot be efficiently updated in-place. To overcome the technology's limitations, SSDs incorporate a software Flash Translation Layer (FTL) that implements out-of-place updates, typically by storing data in a log-structured fashion. Despite a large number of existing FTL algorithms, SSD performance, predictability, and lifetime remain an issue, especially for the write-intensive workloads specific to database applications.In this paper, we show how to design FTLs that are more efficient by using the I/O write skew to guide data placement on flash memory. We model the relationship between data placement and write performance for basic I/O write patterns and detail the most important concepts of writing to flash memory: i) the trade-off between the extra capacity available and write overhead, ii) the benefit of adapting data placement to write skew, iii) the impact of the cleaning policy, and iv) how to estimate the best achievable write performance for a given I/O workload. Based on the findings of the theoretical model, we propose a new principled data placement algorithm that can be incorporated into existing FTLs. We show the benefits of our data placement algorithm when running micro-benchmarks and real database I/O traces: our data placement algorithm reduces write overhead by 20% -75% when compared to state-of-art techniques.

show abstract

Section: State-of-art Data Placement Algorithmsmentioning

confidence: 99%

Section: Data Placement Challengementioning

confidence: 99%

See 1 more Smart Citation

Improving flash write performance by using update frequency

Stoica

Ailamaki

2013

Proc. VLDB Endow.

View full text Add to dashboard Cite

show abstract

“…The merge operation between pages in the log block and original data blocks takes place in the dedicated data group [29]. Chiang et al also proposed data clustering algorithm where data blocks are divided into several regions based on the write access frequency [5]. When a data block is updated, it moves to the upper region.…”

Section: Related Studiesmentioning

confidence: 99%

HFM: Hybrid File Mapping Algorithm for SSD Space Utilization

No¹,

Choi²,

Park³

et al. 2014

Appl. Math. Inf. Sci.

View full text Add to dashboard Cite

Abstract:Although the technology of flash memory is rapidly improving, SSD usage is still limited due to the high cost per storage capacity. An alternative is to construct the hybrid structure where a small SSD partition is combined with the large HDD partition, to expand the file system space to HDD storage capacity while exploiting the performance advantage of SSD. In such a hybrid structure, increasing the space utilization of SSD partition is the critical aspect in generating high I/O performance. In this paper, we present HFM (Hybrid File Mapping) that has been implemented for the hybrid file system integrated with SSD. HFM enables to divide SSD partition into several, logical data sections with each composed of the different extent size. To minimize fragmentation overhead, HFM defines three different ways of partitioning functions based on the extent size of each data section. Furthermore, file allocations on the extent are performed on the partitioned unit (segment) of extents to reuse the remaining free space as much as possible. The experimental result of HFM using three public benchmarks shows that HFM can be effective in increasing the usage of SSD partition and enables to contribute to provide better I/O throughput.

show abstract

“…A page mapping scheme, a fine-grained one, writes all logical pages to anywhere in NAND flash memory [10], which is profitable for many random writes that make small fragmentations. However, it consumes large memory resources to manage the whole page-level mapping information.…”

Section: Flash Translation Layer (Ftl)mentioning

confidence: 99%

A Buffer Management Issue in Designing SSDs for LFSs

Kim

Seol

Maeng

2010

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis letter introduces a buffer management issue in designing SSDs for log-structured file systems (LFSs). We implemented a novel trace-driven SSD simulator in SystemC language, and simulated several SSD architectures with the NILFS2 trace. From the results, we give two major considerations related to the buffer management as follows. (1) The write buffer is used as a buffer not a cache, since all write requests are sequential in NILFS2. (2) For better performance, the main architectural factor is the bus bandwidth, but 332 MHz is enough. Instead, the read buffer makes a key role in performance improvement while caching data. To enhance SSDs, accordingly, it is an effective way to make efficient read buffer management policies, and one of the examples is tracking the valid data zone in NILFS2, which can increase the data hit ratio in read buffers significantly. key words: NAND flash memory, solid state disk, log-structured file system, storage device

show abstract

Using data clustering to improve cleaning performance for flash memory

Cited by 122 publications

References 14 publications

Improving flash write performance by using update frequency

Improving flash write performance by using update frequency

HFM: Hybrid File Mapping Algorithm for SSD Space Utilization

A Buffer Management Issue in Designing SSDs for LFSs

Contact Info

Product

Resources

About