A storage advisor for hybrid-store databases

Abstract

With the SAP HANA database, SAP offers a high-performance in-memory hybrid-store database. Hybrid-store databases-that is, databases supporting row-and column-oriented data management-are getting more and more prominent. While the columnar management offers high-performance capabilities for analyzing large quantities of data, the row-oriented store can handle transactional point queries as well as inserts and updates more efficiently. To effectively take advantage of both stores at the same time the novel question whether to store the given data row-or columnoriented arises. We tackle this problem with a storage advisor tool that supports database administrators at this decision. Our proposed storage advisor recommends the optimal store based on data and query characteristics; its core is a cost model to estimate and compare query execution times for the different stores. Besides a per-table decision, our tool also considers to horizontally and vertically partition the data and manage the partitions on different stores. We evaluated the storage advisor for the use in the SAP HANA database; we show the recommendation quality as well as the benefit of having the data in the optimal store with respect to increased query performance.

Icarus: Towards a multistore database system

“…An example for such a DBMS is SAP HANA [14]. In [15] a storage advisor for SAP HANA is introduced, which recommends the optimal store based on the data and query characteristics. Besides the per-table decision, the tool also considers horizontal and vertical partitioning of the data and the placement of the partitions on different stores.…”

The design of an adaptive column-store system

“…Reference [58] describes on-line physical design tuning for the in-memory database SAP HANA [59]. This database is designed to handle both transactional and analytical workloads.…”

Regularized Cost-Model Oblivious Database Tuning with Reinforcement Learning

“…Table 1 provides a brief classification of related research in automated database configuration, in terms of various dimensions: the offline or online nature of the algorithm (see further), and the physical design aspects being considered by these works (index selection, vertical partitioning, or mixed design together with horizontal partitioning and replication). [11,13,27] Parallel DBs: [2,32] Online [10,24,25,39] [3, 21,36] Offline Algorithms Traditionally, automated database configuration is conducted in an offline manner. In that approach, database administrators (DBAs) identify representative workloads from the trace of historical database queries and updates.…”

“…More recently, as column-oriented databases have attracted a great deal of attention in both academia and industry, online algorithms for automated vertical partitioning becomes critical for such an emerging breed of database systems [3,21,36]. Specifically, a storage advisor for SAP's HANA in-memory database system was proposed in [36] in order to take advantage of both columnar and row-oriented storage layouts. At the core of that storage advisor is a cost model which is used to estimate and compare query execution times for different stores.…”