Availability Models or Mechanisms in Cloud Computing Environment

            Mesbahi, Rahmani, and Hosseinzadeh held a reseach study in 2018 on the topic of  High avaiability in clouds.  The main idea behind this research study is review and research the role and influence of high availability in the cloud computing. For this purpose, researcher conducted a systematic review of the availability in the cloud computing environment (Mesbahi, Rahmani and Hosseinzadeh). Researcher used the mechanism of High availability to reach the findings. The main activities include in the systematic review are identification of need, defining research questions, defining search string, defining research sources, defining exclusion and inclusion criteria, defining data procedure of data extraction, and identification of the primary studies (Mesbahi, Rahmani and Hosseinzadeh).   

            After the complete analysis and research activities researcher concluded results by claiming that cloud outage are the main reasons to financial loss at large scale. Researcher discussed the importance of cloud computing and described how high availability is a biggest challenge for the cloud providers. Furthermore, they also discussed how high availability is essential for the clouds. Considering the importance of high availability cloud provider also seek to ensure HA even in the failure cases (Mesbahi, Rahmani and Hosseinzadeh).

           Researchers said that there are some services that can support improvement in the availability (in cloud computing systems) such as redundancy, load balancing services and checkpointing services. Moreover, researchers also claim that they proposed a High availability solution for the cloud providers that is based on the 3 level layers (Mesbahi, Rahmani and Hosseinzadeh).  

            A research study held by the Endo and research group in 2016 demonstrate that how and what type of uncertiainities can generate negative consequences for cloud computing system. The main idea behind the research study is identification of risk and uncertainty that are usually caused by the challenges associated with the availability, integrity, and confidentiality (Endo, Rodrigues and Gonçalves). The methodology employed by the research study is to collect data and information from the primary and secondary sources.

             Researchers reviewed the uncertainty sources and several types of fundamental approaches that support scheduling in the presence of uncertainties. Furthermore researcher also discussed the methods to mitigate risk factor caused by integrity and availability that are associated with the cloud computing services and loss of information (Endo, Rodrigues and Gonçalves). Mechanism adopted by the research study are availability mechanism, secret sharing mechanism, Homomorphic data encryption, and backup copies mechanism. The research findings conclude that overall cloud computing system get influence from the uncertainty. Therefore there is need to mitigate these risk factors and bring improvement in the service (Endo, Rodrigues and Gonçalves).                     

            The research study held in 2016, by researchers Liang and his research fellows explain how blockchain data provennace can enhacne avaiability and privacy. The main idea behind this research study was to understand the availability mechanism in the cloud computing. Researcher discussed cloud computing along with the block chain based data provenance (Liang, Shetty and Tosh). Researcher used an exploratory methodology and implemented a PorvChain having architecture of blockchain with the objectives such as provenance of data validity, increase in privacy preservation, and real-time cloud data.

             The implementation was made in three main layers data storage layer, provenance database layer, and blockchain network layer. According to the research findings blockchain based data having distributed characteristics increase availability. Moreover, implementation of ProvChain results in the increase of availability that support cloud providers to deal with the risk and challenges associated with the availability (Liang, Shetty and Tosh).        

           Tchernykha, Schwiegelsohn, Talbi, and Babenko and research fellows conducted a research study in 2017 to explain the cloud computing system and avaialbility maechnism. The main idea behind this research study is to investigate the high availability in cloud computing systems. Four categories in roadmap are presented (Tchernykha, Schwiegelsohn and Talbi). Four questions categories are How, When, Where and Which. The mechanism employed in this research study is fault tolerance mechanism, chechpointing mechanism, and High availability mechanism. The results conclude that three really useful solutions for High availability are Monitoring, Detection, and replication. In the research proposed solutions are drawn regarding high availability in cloud computing services. Research conclude that Google cloud is the latest and best cloud computing system.

This study of the Tchernykha, Schwiegelsohn, Talbi, & Babenko, (2016) is significant for the understanding the issues of the uncertainty in various fields particularly those fields which are related to the computational biology for decision making in economics. The phenomena of the uncertainty has been examined by the most of the works according to the perception qualities of the users. It includes the clouds provider’s actions and intentions.

Authors state that a possible technique for resolving the issue of communication and computing imbalance that is related to uncertainty is actually the load balancing that enables improvement in the resource allocation. However, for effective load balancing, it is quite important simplify the notions concerned with the system overload or underload. Additionally, for increasing confidentiality and reliability of the data storing and processing, 6 basic techniques are implemented: homomorphic encryption, regenerating code, erasure code, number system of redundant residue, confidential sharing schemes, and data replication. Authors also explain that the toughest challenge is to find other new approaches.         

According to the research conducted by Melo, Sobrinho, Filho, Feliciano, and Maciel, in 2018 on redundancy mechanism in cloud computing. The main idea behind this research study is analysis of mechanism related to cloud computing system (Melo, Sobrinho and Filho). The methodology employed in the research is sensitive analysis. The mechanism discussed in this research is redundancy mechanism. Results extracted from research findings are based on the solutions and suggestions to bring improvement in the availability. According to the research findings baseline architecture for availability is 0.988571 while the annual time duration is 10 minute for downtime.   

Melo, Sobrinho, Filho, Feliciano, & Maciel, (2018) arguied in this study the sevices which are priovided by the cloud computing system must be accurate according to the various scales as; fault tolerant security, scalability, availability, and so for. The study is also discuss the use of Sensitivity Analysis along with the study of models and modeling hierarchy which are required for the representing redundancy mechanisms. The redundancy mechanisms are considered as the best solution for improving the performance of the cloud computing environments.

Authors describe that sensitivity analysis can definitely help in the determination of significant components for performance and dependability, obtained in an actual analytical model. The methods which are capable of being implemented for analyzing the sensitivity are: PDS or parametric differential analysis, DASI or Discrete Averaged Sensitivity Index, PA or perturbation analysis, regression analysis, correlation analysis, and the factorial design. It is also explained that for further extending the scope, more redundancy mechanisms can be applied as active-active and warm standby models.  They can assist also in incorporating the network availability with various types of topologies.

A research article written by B.Vani and Priya in 2015, is based on the analysis of the avaiability in the cloud computing. The main idea of research study is to determine the availability models in the cloud computing (B.Vani and Priya). Researcher used various types of mechanism to study availability in cloud computing. Research findings conclude that availability is really important in cloud computing. Availability can be enhanced through getting support from the self-healing mechanism. ACR (availability, correlation factor, resources consumption) is really effective for enhance of end-to-end availability. 

 It is explainied in this satudy as  for the distributed systems the Reliability and high availability is considered as the major factors. In cloud computing to providing the highly available and reliable services is known as the significant element for enhancing the satisfaction and confidence which is required to preventing the losses of the revenue. B.Vani & Priya, (2015) has proposed big picture which is divided in to four various steps which can specify through pivotal questions. This study plays significant role for the reference road map.

It is explained that virtualization is actually the development of a virtual model of something like a storage device, operating system, or a hardware platform. Technologies of virtualization seem to decrease hardware and energy costs. Furthermore, they enhance the sharing of resources among the applications which are hosted on several virtual machines. Overall, they are utilized in increasing the manageability of different software systems and reduce the total cost of ownership. The static algorithms which are utilized in load balancing use equal distribution of the traffic among different services. And because of the practical issues, round robin that was weighted was used. It has been concluded that with the emergence of different technological situations, the need for optimizing the availability will also increase.

            In the very first article written by Mesbahi, Rahmani, and Hosseinzadeh, the perfomance is  weak due to the fact that  the approach taken for researching the  subject has no practicality. Only the background of the research topic has been explained and there is no   foundation of the study .  After studying the article, it can be seen that there is nothing  in it regarding the outage use. However, the   research is focusing on other studies and  using them for support which specifies that the recovery has been defined. In the part of the conclusion, challenges have been mentioned and the solution to them has to be determined in other studies. There have been solutions given for solving the issue at hand. The research conducted by Endo and research group is perfect in terms of performance because proper measures have been taken  for researching and literature on the same subject has also been reviewed. The only function of disarray for the second article is the solution that it provides for the current issue. Even though callenges have been mentioned but there is nothing in the context about their solutions.   

The research conducted by Liang and his research fellows is perfect in almost every sense. The performance is remarkable due the practicality of the approach. Other than the basic approach, a proper experimental method has been implemented for the explanation of topic at hand.  The only issue with this research is that not many studies are reviewed in it for supporting the argument that authors are making. Otherwise, challenges have been mentioned along with what should be done for broadening the scope of study.  The research conducted by Tchernykha, Schwiegelsohn, Talbi, and Babenko and research fellows is good in terms of performance due to the fact that researchers havent only relied on other studies for conducting their own research. However, studies haven’t been reveiwed which contributes to the lack of recovery and there is no explanation regarding the outage use and challenges have not been  mentioned in their study. The last two studies practicality of the method is not sufficient enough which makes the performance undefined while the issues and approaches have been explained very well. It makes the study focus on recovery but challenges and uncertainty have not been explained.