Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Building a Columnar Database on RAMCloudData Storage

Building a Columnar Database on RAMCloud: Data Storage [Using RAMCloud with its key-value based data model as storage system for a columnar DBMS results in the question how to map the columnar data to objects in RAMCloud. RAMCloud provides the concept of namespaces. A namespace defines a logical container for a set of objects, where each object key occurs only once. Upon the creation of a new namespace, the parameter server span is set to define how many storage nodes will be used to store the objects of the namespace. These namespaces are assigned to nodes in a round-robin manner. Assignment of key-value pairs across nodes is done by partitioning the range of the hashes of the object keys contained in that namespace.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Building a Columnar Database on RAMCloudData Storage

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/building-a-columnar-database-on-ramcloud-data-storage-SD30ZfsJte

References (0)

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Springer International Publishing
Copyright
© Springer International Publishing Switzerland 2016
ISBN
978-3-319-20710-0
Pages
53 –62
DOI
10.1007/978-3-319-20711-7_4
Publisher site
See Chapter on Publisher Site

Abstract

[Using RAMCloud with its key-value based data model as storage system for a columnar DBMS results in the question how to map the columnar data to objects in RAMCloud. RAMCloud provides the concept of namespaces. A namespace defines a logical container for a set of objects, where each object key occurs only once. Upon the creation of a new namespace, the parameter server span is set to define how many storage nodes will be used to store the objects of the namespace. These namespaces are assigned to nodes in a round-robin manner. Assignment of key-value pairs across nodes is done by partitioning the range of the hashes of the object keys contained in that namespace.]

Published: Jul 8, 2015

Keywords: Block Size; Data Item; Cache Line; Memory Hierarchy; Columnar Data

There are no references for this article.