The uncertain cloud: State of the art and research challenges

Mezni, Haithem; Aridhi, Sabeur; Hadjali, Allel

doi:10.1016/j.ijar.2018.09.009

Cited by 21 publications

(9 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is used as the measure of uncertainty. This uncertainty is caused by noise (coming from the underlying platforms and technologies) and natural uncertainty and variability which is intrinsic to a cloud environment and the Internet [24,25,26]. The CV value (as it is shown in Tables I and II) depends on the workload (the higher the workload, the higher the latency variation) and consistency settings (the stronger the level of consistency, the lower latency variation).…”

Section: A Read/write Latency and Throughput Statisticsmentioning

confidence: 99%

Interplaying Cassandra NoSQL Consistency and Performance: A Benchmarking Approach

Gorbenko

Romanovsky

Tarasyuk

2020

Communications in Computer and Information Science

View full text Add to dashboard Cite

This experience report analyses performance of the Cassandra NoSQL database and studies the fundamental trade-off between data consistency and delays in distributed data storages. The primary focus is on investigating the interplay between the Cassandra performance (response time) and its consistency settings. The paper reports the results of the read and write performance benchmarking for a replicated Cassandra cluster, deployed in the Amazon EC2 Cloud. We present quantitative results showing how different consistency settings affect the Cassandra performance under different workloads. One of our main findings is that it is possible to minimize Cassandra delays and still guarantee the strong data consistency by optimal coordination of consistency settings for both read and write requests. Our experiments show that (i) strong consistency costs up to 25% of performance and (ii) the best setting for strong consistency depends on the ratio of read and write operations. Finally, we generalize our experience by proposing a benchmarking-based methodology for run-time optimization of consistency settings to achieve the maximum Cassandra performance and still guarantee the strong data consistency under mixed workloads.

show abstract

Section: A Read/write Latency and Throughput Statisticsmentioning

confidence: 99%

Interplaying Cassandra NoSQL Consistency and Performance: A Benchmarking Approach

Gorbenko

Romanovsky

Tarasyuk

2020

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…Thus, we model the outbound network latency L to be the overall time consumption of data transmission within two (Google's and Amazon's) cloud data centers (denoted by 2D/B ds ), the data traverse across two corresponding regions' core networks (denoted by 2D/B c ), and the one-way traverse via the metro/edge network (denoted by L m ), as defined in Eq. (1).…”

Section: Modeling Investigationmentioning

confidence: 99%

“…Given the distributed nature of cloud computing, many cloud features are subject to uncertainty to some extent [1]. In particular, the outbound network latency of the cloud can be impacted by various subtleties that are difficult to identify [2], not to mention that the users' lack of infrastructural controllability further increases the challenges in understanding the cloud network performance.…”

Section: Introductionmentioning

confidence: 99%

Characterizing the Cloud's Outbound Network Latency: An Experimental and Modeling Study

Millar-Bilbao

2020

2020 IEEE Cloud Summit

View full text Add to dashboard Cite

Cloud latency has critical influences on the success of cloud applications. Therefore, characterizing cloud network performance is crucial for analyzing and satisfying different latency requirements. By focusing on the cloud's outbound network latency, this case study on Google App Engine confirms the necessity of optimizing application deployment. More importantly, our modeling effort has established a divide-andconquer framework to address the complexity in understanding and investigating the cloud latency.

show abstract

“…Quality of on-premise and off-premise data that generated from variety of source is certainly difficult to control. The data quality control is not as per standard if the generated data exceeds the required value (Chou, 2015;Hazen et al, 2014;Mezni et al, 2018).…”

Section: Quality Controlmentioning

confidence: 99%

Factors Influencing Fog Computing Adoption Based on Quality of Results (QoR) for Heterogeneous Data Analysis: A Proposed Framework

Malic¹,

Izhar²,

Kadir³

2019

IJRTE

View full text Add to dashboard Cite

The rapid increase of data generated has brought challenges on data quality level. Fog computing in general has been supporting the requirements of end user devices that could not be met by cloud computing solution and it is acknowledged to have a major impact on how an organisation decides to adopt for preprocessing a huge amount of data being generated by the devices. Since IoT devices generating very heterogeneous and dynamic data, there are challenges for the level of data quality. The limitation has hindered the development of fog systems framework that capable operating the dynamic execution of edge devices that handling generation and collection large amounts of data on-premise and off-premise. Thus,sufficient operations of identifying Quality of Result enable user to detect any problems when conducting the decision making. The aim of this paper is to address the factors that perceived likely to influence the adoption of fog computing in evaluating the data analysis on data transmitted from the ever increases devices.A conceptual framework has been constructed considering attributes such as heterogeneous data analysis (on-premise and off-premise) and Quality of Results (quality indicators, quality control, validity outcome and reliability outcome).Potential benefits from the implementation of this framework to organisation is it enable to provide greater value and benefits to the business process. The framework of this study could also be influencing and inhibiting the adoption of fog computing.Quality of result has higher chances to satisfy the defined industrial's requirement. In addition, fog-computing adoption is important for serving an environment for industry to execute, monitor, and analyze a large form of data in a fog landscape.

show abstract

The uncertain cloud: State of the art and research challenges

Cited by 21 publications

References 32 publications

Interplaying Cassandra NoSQL Consistency and Performance: A Benchmarking Approach

Interplaying Cassandra NoSQL Consistency and Performance: A Benchmarking Approach

Characterizing the Cloud's Outbound Network Latency: An Experimental and Modeling Study

Factors Influencing Fog Computing Adoption Based on Quality of Results (QoR) for Heterogeneous Data Analysis: A Proposed Framework

Contact Info

Product

Resources

About