Artículos

Fault–Tolerance and Load–Balance Tradeoff in a Distributed Storage System

Estudio de la interdependencia entre tolerancia a fallas y balance de carga en un sistema de almacenamiento distribuido

Moisés Quezada Naquid, Ricardo Marcelín Jiménez and Miguel López Guerrero

Department of Electrical Engineering , Universidad Autónoma Metropolitana – Iztapalapa 09340 – México City, Mexico. E–mail: moises@arte.izt.uam.mx, calu@xanum.uam.mx, milo@xanum.uam.mx

Article received or September 1, 2008.
Accepted on September 25, 2009.

Abstract

In recent years distributed storage systems have been the object of increasing interest by the research community. They promise improvements on information availability, security and integrity. Nevertheless, at this point in time, there is no a predominant approach, but a wide spectrum of proposals in the literature. In this paper we report our findings with a combination of redundancy techniques intended to simultaneously provide fault tolerance and load balance in a small–scale distributed storage system. Based on our analysis, we provide general guidelines for system designers and developers under similar conditions.

Keywords: IDA, storage schemes, failures, recovery, simulations.

Resumen

En los últimos años los sistemas de almacenamiento distribuido han sido objeto de un gran interés por parte de la comunidad de investigadores. Estos sistemas prometen mejoras en cuanto a integridad, seguridad y disponibilidad de la información. Sin embargo, hasta este momento no existe un enfoque predominante, aunque hay diversas propuestas en la literatura. En este artículo reportamos los resultados de nuestras investigaciones con una combinación de técnicas de redundancia que tienen el propósito de proveer simultáneamente tolerancia a fallas y balance de carga en un sistema de almacenamiento distribuido de pequeña escala. Con base en nuestro análisis proporcionamos líneas directrices generales para diseñadores y desarrolladores de sistemas similares.

Palabras clave: IDA, esquemas de almacenamiento, fallas, recuperación, simulaciones.

DESCARGAR ARTÍCULO EN FORMATO PDF

References

1. Adya, A., Bolosky, W. J., Castro, M., Cermark, G., Chaiken, R., Douceur, J. R., Howell, J., Lorch, J. R., Theimer, M. & Wattenhofer R. P. (2002). FARSITE: Federated, Available, and Reliable Storage for an Incompletely Trusted Environment. 5^th Symposium on Operating Systems Design and Implementation (OSDI), Boston, USA, 1–14.         [ Links ]

2. Bhagwan, R., Moore, D., Savage, S. & Voelker, G. M. (2003). Replication strategies for highly available peer–to–peer storage. In Schiper, A., Shvartsman, A.A., Weatherspoon, H., Zhao, B.Y. (Eds.) Future Directions in Distributed Computing (153–158). New Jersey: Springer.         [ Links ]

3. Celeste: An Automatic Storage System (s.f.). Retrieved from http://hub.opensolaris.org/bin/view/Project+celeste/WebHome        [ Links ]

4. Chang, F., Dean, J., Ghemawat, S., Hsieh, W. C., Wallach, D. A., Burrows, M., Chandra, T., Fikes, A. & Gruber, R. E.(2006). Bigtable: A Distributed Storage System for Structured Data. 7^th Symposium on Operating Systems Design and Implementation (OSDI), Seattle, USA, 205–218.         [ Links ]

5. Chen, Y., Edler, J., Goldberg, A. V., Gottlieb, A., Sobti, S. & Yianilos, P. N. (1999). A Prototype implementation of Archival Intermemory. 4th ACM Conference on Digital Libraries, California, USA, 28–37.         [ Links ]

6. Cleversafe (s.f.). Retrieved from http://www.cleversafe.org        [ Links ]

7. Kubiatowicz, J., Bindel, D., Chen, Y., Czerwinski, S., Eaton, P., Geels, D., Gummadi, R., Rhea, S., Weatherspoon, H., Weimer, W., Wells, C. & Zhao, B. (2000) OceanStore: An Architecture for Global–Scale Persistent Storage. ACM SIGPLAN Notice, 35 (11), New York, USA, 190–201.         [ Links ]

8. Marcelín–Jiménez, R., Rajsbaum, S. & Stevens, B. (2006). Cyclic Storage for Fault–tolerant Distributed Executions. IEEE Transactions on Parallel and Distributed Systems, 17(9), 1028–1036.         [ Links ]

9. Rabin, M. O. (1989). Efficient Dispersal of Information for Security, Load Balancing and Fault Tolerance. Journal of the ACM, 38(2), 335–348.         [ Links ]

10. Rodrigues, R. & Liskov, B. (2005). High Availability in DHT's: Erasure Coding vs Replication. Peer–to–Peer Systems IV. 4^th International Workshop on Peer–to–Peer Systems, Lecture Notes on Computer Science , 3640, 226–239.         [ Links ]

11. Rowstron, A. & Druschel, P. (2001). Storage management and caching in PAST, a large–scale, persistent, peer–to–peer storage utility. ACM SIGOPS Operating Systems Review, 35(5), 188–201.         [ Links ]

12. Quezada–Naquid, M., Marcelín–Jiménez, R. & López–Guerrero, M. (2007). Service Policies for a Storage Services Dispatcher in a Distributed Fault–Tolerant Storage Network and their Performance Evaluation. Canadian Conference on Electrical and Computer Engineering (CCECE '07), Vancouver, Canada, 34–40.         [ Links ]

13. OMNeT++: Discrete Event Simulation System (s.f.). Retrieved from http://omnetpp.org/        [ Links ]

14. Yianilos, P. & Sobti, S. (2001). The Evolving Field of Distributed Storage. IEEE Internet Computing. 5 (5), 3539        [ Links ]