4th USENIX Conference on File and Storage TechnologiesAbstract
Pp. 7386 of the Proceedings
Zodiac: Efficient Impact Analysis for Storage Area Networks
Aameek Singh, Georgia Institute of Technology; Madhukar Korupolu and Kaladhar Voruganti, IBM Almaden Research Center
Currently, the fields of impact analysis and policy based management
are two important storage management topics that are
not being treated in an integrated manner. Policy-based storage
management is being adopted by most storage vendors because
it lets system administrators specify high level policies
and moves the complexity of enforcing these policies to the
underlying management software. Similarly, proactive impact
analysis is becoming an important aspect of storage management
because system administrators want to assess the impact
of making a change before actually making it. Impact analysis
is increasingly becoming a complex task when one is dealing
with a large number of devices and workloads. Adding the
policy dimension to impact analysis (that is, what policies are
being violated due to a particular action) makes this problem
even more complex.
In this paper we describe a new framework and a set of optimization
techniques that combine the fields of impact analysis
and policy management. In this framework system administrators
define policies for performance, interoperability, security,
availability, and then proactively assess the impact of desired
changes on both the system observables and policies. Additionally,
the proposed optimizations help to reduce the amount of
data and the number of policies that need to be evaluated. This
improves the response time of impact analysis operations. Finally,
we also propose a new policy classification scheme that
classifies policies based on the algorithms that can be used to
optimize their evaluation. Such a classification is useful in order
to efficiently evaluate user-defined policies. We present an
experimental study that quantitatively analyzes the framework
and algorithms on real life storage area network policies. The
algorithms presented in this paper can be leveraged by existing
impact analysis and policy engine tools.
- View the full text of this paper in HTML and PDF.
Until December 2006, you will need your USENIX membership identification in order to access the full papers. The Proceedings are published as a collective work, © 2005 by the USENIX Association. All Rights Reserved. Rights to individual papers remain with the author or the author's employer. Permission is granted for the noncommercial reproduction of the complete work for educational or research purposes. USENIX acknowledges all trademarks within this paper.
- If you need the latest Adobe Acrobat Reader, you can download it from Adobe's site.