MANAGEMENT METHODS IN SLA-AWARE DISTRIBUTED STORAGE SYSTEMS

Darin Nikolow; Renata Slota; Danilo Lakovic; Paweł Winiarczyk; Marek Pogoda; Jacek Kitowski

doi:10.7494/csci.2012.13.3.35

Authors

Darin Nikolow AGH University of Science and Technology, Faculty of Electrical Engineering, Automatics, Computer Science and Electronics, Department of Computer Science, al. Mickiewicza 30, 30-059 Krakow,
Renata Slota AGH University of Science and Technology, Faculty of Electrical Engineering, Automatics, Computer Science and Electronics, Department of Computer Science, al. Mickiewicza 30, 30-059 Krakow,
Danilo Lakovic AGH University of Science and Technology, Faculty of Electrical Engineering, Automatics, Computer Science and Electronics, Department of Computer Science, al. Mickiewicza 30, 30-059 Krakow
Paweł Winiarczyk AGH University of Science and Technology, ACC Cyfronet AGH, ul. Nawojki 11, 30-950 Krakow,
Marek Pogoda AGH University of Science and Technology, ACC Cyfronet AGH, ul. Nawojki 11, 30-950 Krakow,
Jacek Kitowski AGH University of Science and Technology, Faculty of Electrical Engineering, Automatics, Computer Science and Electronics, Department of Computer Science, al. Mickiewicza 30, 30-059 Krakow, Poland, AGH University of Science and Technology, ACC Cyfronet AGH, ul. Nawojki 11, 30-950 Krakow

DOI:

https://doi.org/10.7494/csci.2012.13.3.35

Keywords:

Service level agreements, mass storage systems, quality of service

Abstract

Traditional data storage systems provide access to user’s data on the “best
effort” basis. While this paradigm is sufficient in many use cases it becomes
an obstacle for applications with Quality of Service (QoS) constraints. Service
Level Agreement (SLA) is a part of the contract agreed between the service
provider and the client and contains a set of well defined QoS requirements
regarding the provided service and the penalties applied in case of violations.
In the paper we propose a set of SLA parameters and QoS metrics relevant
to data storage processes and the management methods necessary for avoiding
SLA violations. A key assumption in the proposed approach is that the underlying
distributed storage system does not provide functionality for resource or
bandwidth reservation for a given client request.