Avaliação dos Serviços de Armazenamento da Amazon Web Services para Gravação e Recuperação de Checkpoints

• Luan Teylo Universidade Federal Fluminense
• Rafaela Brum Universidade Federal Fluminense
• Luciana Arantes Sorbonne Université
• Pierre Sens Sorbonne Université
• Lúcia Drummond Universidade Federal Fluminense

DOI: https://doi.org/10.5753/wtf.2020.12485

Resumo

Os provedores de nuvem oferecem vários serviços para processamento e armazenamento de dados. Alguns desses recursos são propensos a falhas ou revogações. Neste caso, a adoção de técnicas de tolerância a falhas é essencial para minimizar o impacto e o custo que estas falhas podem exercer na execução das aplicações e no orçamento dos usuários. Uma das soluções amplamente adotadas são as técnicas de checkpoint and recovery (CR), que gravam o estado atual da aplicação para que, caso ocorra falhas, esta seja reiniciada a partir do último estado gravado. Assim, essas abordagens exigem um local de armazenamento seguro e disponível ao longo de toda a execução. Felizmente, os serviços de nuvens também podem ser utilizados para este fim, uma vez que diferentes opções de armazenamento também são oferecidos. Neste trabalho, queremos avaliar três desses serviços: o Amazon Simple Storage Service (S3), o Amazon Elastic Block Store (EBS) e o Amazon Elastic File System (EFS). O objetivo é caracterizar e avaliar o desempenho desses serviços em relação a gravação e recuperação de checkpoints no contexto de falhas induzidas pela revogação de máquinas virtuais.

Palavras-chave: checkpoint e recovery, serviços de armazenamento, avaliação

Referências

Alves, M. M. and Drummond, L. M. (2017). A multivariate and quantitative model forpredicting cross-application interference in virtual environments. Journal of Systemsand Software, 128:150 — 163.

AWS (2020a). Amazon EC2 pricing. https://aws.amazon.com/ec2/pricing/. Accessed 15 March 2020.

AWS (2020b). Amazon Elastic Block Store. https://aws.amazon.com/ebs/.Accessed 15 March 2020.

AWS (2020c). Amazon S3. https: //aws.amazon.com/s3/. Accessed 15 March2020.

Di, S., Robert, Y., Vivien, F., Kondo, D., Wang, C.-L., and Cappello, F. (2013). Optimi-zation of cloud task processing with checkpoint-restart mechanism. In Proceedings ofthe International Conference on High Performance Computing, Networking, Storageand Analysis, SC "13, New York, NY, USA. Association for Computing Machinery.

Egwutuoha, I. P., Levy, D., Selic, B., and Chen, S. (2013). A survey of fault tolerancemechanisms and checkpoint/restart implementations for high performance computingsystems. The Journal of Supercomputing, 65(3):1302-1326.

Elnozahy, E. N., Alvisi, L., Wang, Y.-M., and Johnson, D. B. (2002). A survey of rollback-recovery protocols in message-passing systems. ACM Computing Surveys (CSUR),34(3):375-408.

EMELYANOV, P. (2011). Criu: Checkpoint/restore in userspace, july 2011. URL:https://criu. org.

Hargrove, P. H. and Duell, J. C. (2006). Berkeley lab checkpoint/restart (BLCR) for linuxclusters. Journal of Physics: Conference Series, 46:494-499.

Meroufel, B. and Belalem, G. (2018). Optimization of checkpointing/recovery strategy incloud computing with adaptive storage management. Concurrency and Computation:Practice and Experience, 30(24):e4906. e4906 CPE-17-0585.R2.

Nicolae, B. and Cappello, F. (2011). Blobcr: Efficient checkpoint-restart for hpc applicati-ons on iaas clouds using virtual disk image snapshots. In SC "11: Proceedings of 2011International Conference for High Performance Computing, Networking, Storage andAnalysis, pages 1-12, Seatle, WA, USA. IEEE.

Poola, D., Ramamohanarao, K., and Buyya, R. (2014). Fault-tolerant workflow schedu-ling using spot instances on clouds. Procedia Computer Science, 29:523 — 533. 2014International Conference on Computational Science.

Rizun, R. (2011). S3fs: Fuse-based file system backed by amazon s3. https://github.com/s3fs-fuse/s3fs-fuse/.

Ruiz-Alvarez, A. and Humphrey, M. (2011). An automated approach to cloud storageservice selection. In Proceedings of the 2nd International Workshop on ScientificCloud Computing, ScienceCloud "11, page 39-48, New York, NY, USA. Associationfor Computing Machinery.

Services, A. W. (2020a). Amazon Elastic File System. https: //aws.amazon.com/efs/. Accessed 07 April 2020.

Services, A. W. (2020b). Boto 3 Documentation. https://boto3.amazonaws.com/vl/documentation/api/latest/index.html. Accessed 01 April2020.

Services, A. W. (2020c). Cloud Storage with AWS. https://aws.amazon.com/products/storage/?ncl=h 1s. Accessed 02 April 2020.

Teylo, L., Arantes, L., Sens, P., and de A. Drummond, L. M. (2019). A hibernation awaredynamic scheduler for cloud environments. In Proceedings of the 48th InternationalConference on Parallel Processing: Workshops, ICPP 2019, New York, NY, USA.Association for Computing Machinery.

Yaothanee, J. and Chanchio, K. (2019). An in-memory checkpoint-restart mechanism fora cluster of virtual machines. In 2019 16th International Joint Conference on Compu-ter Science and Software Engineering (JCSSE), pages 131-136, Chonburi, Thailand,Thailand. IEEE.

Yi, S., Kondo, D., and Andrzejak, A. (2010). Reducing costs of spot instances via check-pointing in the amazon elastic compute cloud. In 2010 IEEE 3rd International Confe-rence on Cloud Computing, pages 236-243, Miami, FL, USA. IEEE, IEEE.

Zhou, A., Sun, Q., and Li, J. (2017). Enhancing reliability via checkpointing in cloudcomputing systems. China Communications, 14(7):1-10.