Home // DBKDA 2012, The Fourth International Conference on Advances in Databases, Knowledge, and Data Applications // View article

The Multi-Tenant Data Placement Problem

Authors:
Jan Schaffner
Dean Jacobs
Tim Kraska
Hasso Plattner

Keywords: multi-tenancy; software-as-a-service; databases; enterprise applications; column-store

Abstract:
With the advent of the Software-as-a-Service (SaaS) deployment model, managing operational costs becomes more and more important for providers of hosted software. The cost for hosting and providing a service is directly proportional to the operational margin that can be achieved when running a SaaS business. Possible avenues for reducing operational costs are consolidation (i.e. co-locating multiple customers onto the same server) and automation of cluster management (i.e. migration of customers between servers, automatic replication for performance or high availability). In this paper, we propose the formalization for the problem of assigning "tenants" (i.e. the customers) to servers of an on-demand database cluster. We will pose this problem in the form of an optimization problem, omitting database specifics and thus presenting the problem in an abstract fashion, using the metaphor of assigning tokens to baskets (i.e., tenants to servers). This formalization is both a first step towards solving the placement problem in an efficient way and a helpful basis for comparing multiple solutions to the problem to each other.

Pages: 157 to 162

Copyright: Copyright (c) IARIA, 2012

Publication date: February 29, 2012

Published in: conference

ISSN: 2308-4332

ISBN: 978-1-61208-185-4

Location: Saint Gilles, Reunion

Dates: from February 29, 2012 to March 5, 2012