A case for flash memory ssd in enterprise database applications

Abstract

Due to its superiority such as low access latency, low energy consumption, light weight, and shock resistance, the success of flash memory as a storage alternative for mobile computing devices has been steadily expanded into personal computer and enterprise server markets with ever increasing capacity of its storage. However, since flash memory exhibits poor performance for small-to-moderate sized writes requested in a random order, existing database systems may not be able to take full advantage of flash memory without elaborate flash-aware data structures and algorithms. The objective of this work is to understand the applicability and potential impact that flash memory SSD (Solid State Drive) has for certain type of storage spaces of a database server where sequential writes and random reads are prevalent. We show empirically that up to more than an order of magnitude improvement can be achieved in transaction processing by replacing magnetic disk with flash memory SSD for transaction log, rollback segments, and temporary table spaces.

WPCB-Tree: A Novel Flash-Aware B-Tree Index Using a Write Pattern Converter

“…The main goals of this WPCB-tree are as follows: (1) to quickly create the index structure with very little use of the main memory resource; (2) to alleviate the disadvantage of the random write pattern by converting the write pattern from random to sequential; and (3) to ensure that the data is nonvolatile in the case of sudden power failure. The WPCB-tree avoids the erase-before-write characteristic and reduces the number of write operations by using a novel method and mechanism.…”

“…Although these advantages have lead flash memory to become a commonly used memory technology, its downsides, including erase-before-write, limited life cycle, and higher cost than a normal HDD, should not be overlooked [1][2][3][4][5]. Among these mentioned weak points, erase-before-write is a critical characteristic that needs to be mitigated because an erase operation leads to many read and write operations and shortens the life cycle of flash memory.…”

An genetic algorithm approach for profiling computational performance measures

“…Changes in the hardware can also improve performance, since applications like large databases are usually IO intensive. Solid-state Drives(SSD) in general have a better performance then Hard Disk Drivers(HDDs) for this scenario (Lee et al 2008) …”

Memory Compaction Scheme with Block-Level Buffer for Large Flash Memory

“…Block-level buffer management policy not only invokes relatively fewer merge operations than page-level buffer management policy but also invokes switch merge or partial merge rather than full merge for merge operation. Block-level buffer management policy shows better overall performance than the page-level buffer management policy [12], [16]- [18].…”