Analysis of a Virtual Storage Service Exploring Classes of Objects in the Cloud and Access Patterns

• Samuel de Oliveira Ribeiro Universidade Federal do Piauí
• Glauber Dias Gonçalves Universidade Federal do Piauí http://orcid.org/0000-0002-0742-4346
• Francisco Airton Silva Universidade Federal do Piauí http://orcid.org/0000-0002-8211-6060
• Alex Vieira Universidade Federal de Juiz de Fora http://orcid.org/0000-0003-0821-126X
• Jussara Almeida DCC-UFMG http://orcid.org/0000-0001-9142-2919

DOI: https://doi.org/10.5753/wperformance.2020.11108

Abstract

A virtual storage service uses third-party cloud infrastructure to store data as objects of different classes and costs. In this article, we investigate a key aspect of this service: the prediction of the frequency of accesses to the data to allocate them in appropriate storage classes. predicting the frequency of data access to properly store data in different storage classes of the cloud infrastructure. Based on data access patterns of users, we propose a framework that dynamically predicts appropriate classes to store data (objects) in the near future. We evaluated the performance of this framework through trace-based data access simulations from Dropbox. Our results show high performance for predictions, which leads to savings of up to 14% in the storage costs.

Keywords: storage, access, prediction, machine learning, cloud

References

Bocchi, E., Drago, I., and Mellia, M. (2015). Personal Cloud Storage: Usage, Performance and Impact of Terminals. In Proc. of the IEEE CloudNet.

Cisco (2019). Cisco Global Cloud Index: Forecast and Methodology, 2016–2021 WhitePaper. Disponível em https://www.cisco.com- Document ID 1513879861264127.

Gonçalves, G. et al. (2016). The impact of content sharing on cloud storage bandwidth consumption. IEEE Internet Computing, 20(4):26–35.

Gracia-Tinedo, R., García-López, P., Gómez, A., and Illana, A. (2016). Understanding data sharing in private personal clouds. In Proc. of the IEEE CLOUD.

Hsu, Y., Irie, R., Murata, S., and Matsuoka, M. (2018). A novel automated cloud storage tiering system through hot-cold data classification. In Proc. of the IEEE CLOUD.

Irie, R., Murata, S., Hsu, Y., and Matsuoka, M. (2018). A novel automated tiered storage architecture for achieving both cost saving and qoe. In Proc. of the IEEE SC2.

Josep, A. D., Katz, R., Konwinski, A., Gunho, L., PattErson, D., and Rabkin, A. (2010). A view of cloud computing. Communications of the ACM, 53(4).

Kaushik, R. T. and Bhandarkar, M. (2010). Greenhdfs: towards an energy-conserving, storage-efficient, hybrid hadoop compute cluster. In Proc. of the USENIX.

Liu, M., Pan, L., and Liu, S. (2019). To transfer or not: An online cost optimization algorithm for using two-tier storage-as-a-service clouds. IEEE Access, 7:94263–94275.

Mesnier, M., Ganger, G. R., and Riedel, E. (2003). Object-based storage.IEEE Communications Magazine, 41(8):84–90.

Muralidhar, S. et al. (2014). f4: Facebook’s warm blob storage system. In Proc. of the OSDI.

Ribeiro, S. O. et al. (2019). Análise de um serviço virtual de armazenamento em nuvem ciente de padrões de acesso por usuários. In Proc. of the ERCEMAPI.