@MISC{Kimmigy_basicbuilding, author = {Daniel Kimmigy and Reimar Hofmann}, title = {Basic Building Blocks for Column-Stores}, year = {} }
Share
OpenURL
Abstract
Abstract—A constantly increasing CPU-memory gap as well as steady growth of main memory capacities have increased interest in column store systems due to potential performance gains within the realm of database solutions. In the past, several monolithic systems have reached maturity in the commercial and academic spaces. However, a framework of low-level and modular components for rapidly building column store based applications has yet to emerge. A possible field of application is the rapid development of high-performance components in various data-intensive areas such as text-retrieval systems and recommenda-tion systems. The main contribution of this paper is a column-store-tool-kit, a basic building block of low-level components for constructing applications based on column store principles. We present a minimal amount of necessary structural elements and associated operations required for building applications based on our column-store-kit. The eligibility of our toolkit is demonstrated subsequently in using the components of our toolkit for building different query execution plans. This part of work is a first step in our effort for the construction of a pure colmun-store based query optimizer. Keywords—Column store; basic components; framework; rapid prototyping; TPC-H benchmark; query-optimizer; query-execution plan. I.