Cloud or In-House Service? Strategic Joining and Social Optimality in Hybrid Service Systems With Retrial Orbit

Zhu, Sheng; Wang, Jinting; Li, Wei Wayne

doi:10.1109/jsyst.2023.3246727

Cited by 5 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many works have been published about the queueing theory, such as Ref. [19][20][21][22] , and among others. Goswami et al [19] constructed a queueing model with finite multiservers, which controlled the number of virtual machines in real-time by cloud architecture.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Recently, the strategic behavior of customers has evoked the interest of many researchers. Zhu et al [22] proposed a two-tier cloud service system with a queue-length-based admission control mechanism and analyzed the equilibrium joining probability of customers' behavioral strategies. Interested readers can refer to Ref.…”

Section: Literature Reviewmentioning

confidence: 99%

“…We derive various performance measures based on the steady-state probability distribution, such as the mean number of jobs in orbit space, N orbit , can be expressed as Equation ( 14): LOS 2 -control strategy. The hybrid cloud system with LOS 2control strategy has been studied by Zhu et al [22] . They derived relevant performance measures, such as the probability of having i jobs/requests in the private cloud, the mean waiting time of a job newly arriving in the private the mean number of jobs in the orbit, the mean waiting time of jobs in the orbit and so on.…”

Section: /11mentioning

confidence: 99%

“…They derived relevant performance measures, such as the probability of having i jobs/requests in the private cloud, the mean waiting time of a job newly arriving in the private the mean number of jobs in the orbit, the mean waiting time of jobs in the orbit and so on. Based on these performance measures of the system model, Zhu et al [22] further explored the conditional equilibrium joining probability of entering the private cloud when the public cloud is not open, and they determined the cooperatively optimal retrial rate and the noncooperatively optimal (L,q)-controletrial rate for a given queue-length information.…”

Section: /11mentioning

confidence: 99%

See 3 more Smart Citations

Research on Data Risk Control Strategies for Hybrid Cloud

Zhao,

Zhu

2024

J Inform Anal

View full text Add to dashboard Cite

Objective: Hybrid cloud provides an efficient and relatively secure service, widely used in commercial and public affairs. There are already several specific risk control strategies for hybrid clouds in the existing cloud computing literature, but there is still a need to design data risk control strategies that cater to users with different risk preferences. Methods: Firstly, by using a classification method, we propose four data risk control strategies that are suitable for different cloud environments. Then, we use the queueing theory to model the hybrid cloud system and derive some relevant indicators of system performance. Results: We propose four new risk control strategies for the personalized needs of enterprises in different scenarios, namely, L-control strategy, LOS1-control strategy, LOS2-control strategy, and (L,q)-control strategy. Conclusions: It is more practical to use a hybrid cloud system for the protection of data security. This work proposes four risk control strategies for different situations to help companies with risk control, namely, L-control strategy, LOS1-control strategy, LOS2-control strategy, and (L,q)-control strategy. These risk control strategies provide theoretical assistance for exploring the allocation of cloud resources.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: /11mentioning

confidence: 99%

Section: /11mentioning

confidence: 99%

See 2 more Smart Citations

Research on Data Risk Control Strategies for Hybrid Cloud

Zhao,

Zhu

2024

J Inform Anal

View full text Add to dashboard Cite

show abstract

“…Furthermore, in a hybrid cloud system, there exists a two-tier model where web service companies requiring computing resources allocate some of their upcoming work to the cloud and the remainder to their on-premise subsystems to avail services. Considering the strategic behavior of customers, Zhu et al [23] studied the optimal control in this hybrid cloud system with a two-tier mechanism. They analyzed customers' behavior strategies and obtained the optimal admission control approach to maximize the server profit and social welfare.…”

Section: Literature Reviewmentioning

confidence: 99%

Optimal ordering strategy and budget allocation for the Covid-19 vaccination planning

Liu,

Zhu,

Wang

2024

Math. Model. Nat. Phenom.

Self Cite

View full text Add to dashboard Cite

Abstract. During the COVID-19 pandemic, the most important thing was to control the overall infection rate. To achieve this goal, social managers can choose to use vaccines with different production cycles and therapeutic effects for epidemic prevention and control under financial budget constraints. We adopt a two-tier queueing system with reneging to characterize the operation management of COVID19 vaccine ordering and vaccination, in which a higher-efficacy vaccine queue (HQ) and a lower-efficacy vaccine queue (LQ) are employed to account for two types of vaccines service. In light of this framework, a recursive formula is proposed for deriving the infection rates of residents in both HQ and LQ. Social managers can achieve the lowest total infection rate by selecting appropriate vaccine ordering strategies under fixed service capacity, or by allocating financial budgets reasonably under the investment cost regime. Accordingly, we obtain socially optimal vaccine ordering strategies and financial budget allocation. Finally, we analyze the sensitivity of various parameters to relevant optimal strategies and discover that utilizing a mixed ordering strategy is socially optimal in most circumstances. However, in some extreme cases, ordering a single type of vaccine (higher- or lower-efficacy) may result in the lowest societal infection rate.

show abstract