Loading...
Please wait, while we are loading the content...
Similar Documents
Poster : Risk-based Optimization of Resource Provisioning in Mobile Edge Computing
| Content Provider | Semantic Scholar |
|---|---|
| Author | Badri, Hossein Grosu, Daniel Bahreini, Tayebeh |
| Copyright Year | 2018 |
| Abstract | Resource provisioning is a challenging issue for Mobile Edge Computing (MEC) service providers significantly impacting the efficiency of the system and the Quality of Service (QoS). This is due to the existence of nondeterministic parameters which makes it difficult to manage the resources efficiently. Researches have addressed the resource provisioning problem in Cloud Computing (CC) considering the uncertainty of parameters. Maguluri et al. [3] introduced a stochastic model for load balancing and scheduling in CC clusters, where the arrival time and duration of jobs are stochastic. Chaisiri et al. [2] proposed a stochastic model for the cloud resource provisioning problem under uncertainty of resource prices and demands. Wang et al. [5] developed a model for mapping virtual machines into cloud servers assuming that the completion time of the requests of users is stochastic. Compared to CC, resource provisioning in MEC is expected to be more challenging. First, resource requirements of mobile applications are unknown prior to running applications on servers, and second, edge servers have more restricted capacity than the cloud servers. In this paper, we propose a riskedbased optimization approach to resource provisioning in MEC systems with the aim of taking into account the risk of overloading of edge servers when making allocation decisions. Assuming that resource requirements of mobile applications are stochastic parameters, we formulate the problem as a chance-constrained stochastic program. In order to solve the problem in reasonable amount of time, we employ the Sample Average Approximation (SAA) method [1]. We evaluate the efficiency of the proposed approach by conducting an experimental analysis on instances with different problem settings. Our contributions are as follows: (i) We propose a riskbased optimization approach to resource provisioning problem in MEC; (ii) We propose a clustering-based approach to approximate the probability distributions of resource requirements of mobile applications; (iii) We propose the use of the SAA method to solve the chance-constrained stochastic program; and (iv) We provide a comprehensive analysis of the effects of the overloading risk factor on the utilization rates of servers and the QoS. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://acm-ieee-sec.org/2018/posters/SEC18_Poster_Badri.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Poster |
