An automated implementation of hybrid cloud for performance evaluation of distributed databases

Mansouri, Yaser; Prokhorenko, Victor; Babar, Muhammad Ali

doi:10.1016/j.jnca.2020.102740

Cited by 39 publications

(30 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To build a secure and cost-free connection between OpenStack and Azure, we opted to use WireGuard [19] which is a Linux kernel-based VPN. WireGuard is userfriendly and offers several benefits such as security, reliability in connection, zero-cost and high throughput [20].…”

Section: B Cloud Connectivitymentioning

confidence: 99%

“…Here, we outline the steps we followed for setting up a hybrid cloud. The implementation, inspired from [20], consists of 8 steps as shown in Figure 1. Step 8 -Data Routing: This step implemented data routing in Azure and OpenStack.…”

Section: Hybrid Cloud Implementationmentioning

confidence: 99%

“…Hybrid cloud implementation and cloud bursting: Several studies (e.g., [7], [20], [26]- [28]) have explored the implementation of the hybrid cloud and cloud bursting. Mansouri et al [20] used WireGuard [19] to connect private and public clouds for implementing a hybrid cloud to compare the impact of cloud bursting on the performance of six distributed databases (e.g., Cassandra and MongoDB). Clemente-Castelló et al [27] leveraged a low throughput and high latency link to connect two clouds hosted on OpenStack for building a collaborative hybrid cloud.…”

Section: Confirm Version Compatibilitymentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Distributed Data Processing Frameworks in Hybrid Clouds

Ullah¹,

Dhingra²,

Xia³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Distributed data processing frameworks (e.g., Hadoop, Spark, and Flink) are widely used to distribute data among computing nodes of a cloud. Recently, there have been increasing efforts aimed at evaluating the performance of distributed data processing frameworks hosted in private and public clouds. However, there is a paucity of research on evaluating the performance of these frameworks hosted in a hybrid cloud, which is an emerging cloud model that integrates private and public clouds to use the best of both worlds. Therefore, in this paper, we evaluate the performance of Hadoop, Spark, and Flink in a hybrid cloud in terms of execution time, resource utilization, horizontal scalability, vertical scalability, and cost. For this study, our hybrid cloud consists of OpenStack (private cloud) and MS Azure (public cloud). We use both batch and iterative workloads for the evaluation. Our results show that in a hybrid cloud (i) the execution time increases as more nodes are borrowed by the private cloud from the public cloud, (ii) Flink outperforms Spark, which in turn outperforms Hadoop in terms of execution time, (iii) Hadoop transfers the largest amount of data among the nodes during the workload execution while Spark transfers the least amount of data, (iv) all three frameworks horizontally scale better as compared to vertical scaling, and (v) Spark is found to be least expensive in terms of $ cost for data processing while Hadoop is found the most expensive.

show abstract

Section: B Cloud Connectivitymentioning

confidence: 99%

Section: Hybrid Cloud Implementationmentioning

confidence: 99%

Section: Confirm Version Compatibilitymentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of Distributed Data Processing Frameworks in Hybrid Clouds

Ullah¹,

Dhingra²,

Xia³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In terms of innovation recently, the library uses big data to enrich the library database, improve the capacity of librarians, facilitate inter-library loan and provide personalized knowledge services (Li et al , 2017). In this case, NoSQL databases are often the best solutions for modern big data applications (Mansouri et al , 2020). Many scholars so far researched to know about the preferring attitude of e-resources.…”

Section: Literature Reviewmentioning

confidence: 99%

Knowledge-based society and emerging disciplines: a correlation of academic performance

Rafi

Ahmad

Khan

et al. 2020

View full text Add to dashboard Cite

Purpose Digital libraries promote and accelerate scientific research in academic institutions. The subscribed database resources of digital libraries have become an increasingly valuable asset for researchers. Database resources help generate new ideas, determine research directions and promote productive academic interaction between teachers and students in the information age. The purpose of this study is to examine the use of electronic resources by students in various databases, the research productivity of the faculty in the science network and the number of students who graduate each year. Design/methodology/approach This study uses a quantitative method to collect secondary data from the central database of the Higher Education Commission (HEC) for the population of 26 universities for 2 years (2015–2016). In addition to the HEC digital library, data was also collected from the Web of Science to determine the quality academic performance of faculty and researchers. Moreover, in the study, the total strength of teaching staff and doctoral faculty was extracted from the HEC website for investigation. The authors applied the Spearman’s correlation test to the secondary data using Statistical Package for Social Sciences version 25. Findings The correlation results of the enrolled students and the downloaded papers from various databases were statistically insignificant (p > 0.05). However, the result showed a positive correlation (p < 0.05) between the use of selected/known databases from a number of databases accessed by the HEC. More importantly, it turns out that the faculty’s productivity in the scientific network and the number of students who graduated from public and private universities are found to be insignificant (p > 0.05). However, the authors found a positive correlation (p < 0.05) between doctoral and non-doctoral faculties, which show that a significant number of non-doctoral faculties are still actively involved in teaching and research. Originality/value Research based on academic activities by faculties and students, performed for the first time on the basis of secondary data, will help the HEC and university management to determine the right direction and develop plans to improve academic performance and research quality.

show abstract

“…O uso de nuvens para execução de aplicações HPC pode ser visto como um bom complemento para estruturas locais de supercomputadores e clusters, não podendo, ainda, substituílas totalmente. Abordagens de utilização de nuvens híbridas, nas quais ocorre uma integração entre infraestruturas de nuvens públicas e privadas, permitem a ocorrência de cloud bursting [11]. Isto faz com que aplicativos utilizem toda a capacidade dos recursos computacionais de uma nuvem privada e, em reação a este aumento, migurem suas tarefas em ambientes de nuvem pública, à medida que os recursos locais se tornem escassos.…”

Section: Comounclassified

Cloud infrastructure for HPC investment analysis

Santos

Cavalheiro

2020

RITA

View full text Add to dashboard Cite

With the consolidation of cloud computing technology, there is a growing interest in exploring it to support High Performance Computing (HPC). However, migrating such applications to public or private cloud environments brings some challenges, in particular, the cost in financing the migration process. In this paper, a literature review is presented with selected papers about analyzing cloud infrastructure investments. In particular, the selected papers analyse how investments impact applications. For discussion of related works, conditions for running HPC applications in the cloud are characterized.

show abstract

An automated implementation of hybrid cloud for performance evaluation of distributed databases

Cited by 39 publications

References 31 publications

Evaluation of Distributed Data Processing Frameworks in Hybrid Clouds

Evaluation of Distributed Data Processing Frameworks in Hybrid Clouds

Knowledge-based society and emerging disciplines: a correlation of academic performance

Cloud infrastructure for HPC investment analysis

Contact Info

Product

Resources

About