An efficient B-tree layer implementation for flash-memory storage systems

Wu, Chin-Hsien; Kuo, Tei-Wei; Ch’ang, Lan-Yang

doi:10.1145/1275986.1275991

Cited by 151 publications

(83 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our cost estimation considers the asymmetry of read and write speeds of the flash SSD, as well as the different patterns of sequential and random accesses. Subsequently, we analytically compare the costs of FD-tree with the representative B+-tree variants including the standard B+-tree [8], the LSM-tree [24], and BFTL [28]. Given n index entries, the search cost of FD-tree is close to that of B+-tree, and matches the optimal search cost O(logBn) I/Os, where B is the page size.…”

Section: Introductionmentioning

confidence: 93%

“…Due to the poor random write performance of flash SSDs, write optimized tree indexes [28,23] have been proposed to improve the update performance. BFTL [28] was proposed to balance the inferior random write performance and fast random read performance for flash memory based sensor nodes and embedded systems.…”

Section: Write Optimized Tree Indexingmentioning

confidence: 99%

“…BFTL [28] was proposed to balance the inferior random write performance and fast random read performance for flash memory based sensor nodes and embedded systems. It allows the index entries in one logical B-tree node to span over multiple physical pages, and maintains an in-memory table to map each B-tree node to multiple physical pages.…”

Section: Write Optimized Tree Indexingmentioning

confidence: 99%

“…We compare the I/O cost of FD-tree with other B+-tree variants including the standard B+-tree [8], LSM-tree [24] and BFTL [28]. Table 2 shows their costs on the search and insertion.…”

Section: Cost Analysis and Cost Modelmentioning

confidence: 99%

“…However, all these indexes are suboptimal on search performance. Recently, there has been initial work, in particular BFTL [28], on optimizing the B+-tree for flash memory in embedded systems. Unfortunately, BFTL improves the update performance at the expense of deteriorated search performance.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Tree indexing on solid state drives

Yang

et al. 2010

Proc. VLDB Endow.

142

View full text Add to dashboard Cite

Large flash disks, or solid state drives (SSDs), have become an attractive alternative to magnetic hard disks, due to their high random read performance, low energy consumption and other features. However, writes, especially small random writes, on flash disks are inherently much slower than reads because of the erase-beforewrite mechanism.To address this asymmetry of read-write speeds in tree indexing on the flash disk, we propose FD-tree, a tree index designed with the logarithmic method and fractional cascading techniques. With the logarithmic method, an FD-tree consists of the head tree -a small B+-tree on the top, and a few levels of sorted runs of increasing sizes at the bottom. This design is write-optimized for the flash disk; in particular, an index search will potentially go through more levels or visit more nodes, but random writes are limited to a small area -the head tree, and are subsequently transformed into sequential ones through merging into the lower runs. With the fractional cascading technique, we store pointers, called fences, in lower level runs to speed up the search. Given an FD-tree of n entries, we analytically show that it performs an update in O(log B n) sequential I/Os and completes a search in O(log B n) random I/Os, where B is the flash page size. We evaluate FD-tree in comparison with representative B+-tree variants under a variety of workloads on three commodity flash SSDs. Our results show that FD-tree has a similar search performance to the standard B+-tree, and a similar update performance to the write-optimized B+-tree variant. As a result, FD-tree dominates the other B+-tree index variants on the overall performance on flash disks as well as on magnetic disks.

show abstract

Section: Introductionmentioning

confidence: 93%

Section: Write Optimized Tree Indexingmentioning

confidence: 99%

Section: Write Optimized Tree Indexingmentioning

confidence: 99%

“…We compare the I/O cost of FD-tree with other B+-tree variants including the standard B+-tree [8], LSM-tree [24] and BFTL [28]. Table 2 shows their costs on the search and insertion.…”

Section: Cost Analysis and Cost Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations