Performance Improvement of Database Compression for OLTP Workloads

Abstract

SUMMARYAs data volumes explode, data storage costs become a large fraction of total IT costs. We can reduce the costs substantially by using compression. However, it is generally known that database compression is not suitable for write-intensive workloads. In this paper, we provide a comprehensive solution to improve the performance of compressed databases for write-intensive OLTP workloads. We find that storing data too densely in compressed pages incurs many future page splits, which require exclusive locks. In order to avoid lock contention, we reduce page splits by sacrificing a couple of percent of space savings. We reserve enough space in each compressed page for future updates of records and prevent page merges that are prone to incur page splits in the near future. The experimental results using TPC-C benchmark and MySQL/InnoDB show that our method gives 1.5 times higher throughput with 33% space savings compared with the uncompressed counterpart and 1.8 times higher throughput with only 1% more space compared with the state-of-the-art compression method developed by Facebook.

Impact of working capital on firm performance: DoesITmatter?

“…Therefore, for the first instrument for IT infrastructure, following Loughran and Ritter (2004), we created a dummy variable for technology firms (High IT = 1) that traditionally invest heavily in IT infrastructure 1 . Our second instrument for IT infrastructure was firm‐level “hardware investment,” which has been used as an instrument for IT in prior studies that argue that although “hardware investment is highly correlated with IT investment, it does not directly influence firm performance (Rai et al, 1997) except through the effect of IT investments on performance (Lee & Mithas, 2014, p. 9).” We conducted a Sargan test to statistically validate instrument relevance and found both instruments to be valid ( p > 0.1).…”

A Selective Compression Strategy for Performance Improvement of Database Compression

“…Recently, Lee [1] has improved the performance of database compression for OLTP workloads by avoiding lock contention. These techniques need modification of DBMS engines.…”

“…We define the space savings (SS) according to literature [2,8] in Formula (1). According to Lee [1], if data is too densely stored in compressed pages, page splits will be often incurred by updates. Page splitting lock the whole index in exclusive mode and lead to low concurrency.…”

“…However, it is generally known that database compression is not suitable for write-intensive workloads like OLTP [1,2]. The main reason is that updates of compressed pages incur many page splits, which require exclusive locks [1]. Furthermore, updates compressed page size.…”