بررسی برنامه های کاربردی محاسبات ابری مبتنی بر عامل
Abstract
1- Introduction
2- Different approaches to classifying existing research
3- A cloud-based approach to classify the relationship between CC and MAS
4- Analysis of open issues and opportunities
5- Conclusions
References
Abstract
In the state of the art, there are very few studies on agent-based Cloud Computing. Nevertheless, this is an emerging trend and the number of studies and applications in this field is beginning to increase. Cloud Computing and Agents are complementary technologies. The features of Cloud Computing can provide advanced computational characteristics to multi-agent systems. In turn, the inclusion of agent systems in the core of the Cloud platform makes it possible to incorporate different functionalities, such as reasoning and learning capabilities. This study analyzes the emerging relationship between both distributed systems. Specifically, this study proposes a new classification from the point of view of Cloud Computing, based on the reference architecture proposed by the National Institute of Standards and Technology and the different responsibilities of each of the roles that participate in the Cloud Computing paradigm as identified in the architecture: Provider, Consumer, Broker, Carrier and Auditor.
Introduction
The technology industry and the scientific community have made great strides in recent years towards implementing the Cloud Computing (CC) paradigm. This has resulted in a rapid growth of both private and public platforms [1–4] aimed at providing innovative solutions that can resolve the current needs of the CC paradigm. There is no doubt that the general social approval of this paradigm [5] has in large part led to its quick development, in addition to the economic interests of large technology companies which have focused on advancing the purely technical aspects [6,7]. From an external point of view, a CC platform offers three well-known types of services (software, platform and infrastructure) [8], but today the literature already refers to Something as a Service (XaaS) [9–11], assuming that any type of computer service offered over the Internet can be of the Cloud type, which makes it possible to include new services such as FaaS containers (Function as a Service) [12], AaaS (Analytics as a Service) [13], or even HaaS (Human as a Service) [14]. The novelty that distinguishes this technology from previous technologies is its ability to offer any type of computing capacity as a service through the Internet. The marketing model used in the CC paradigm is also innovative, as it is based on a pay-as-you-go concept [6], in which users must negotiate the terms of the Service Level Agreement (SLA) in order to access services [15]. Once this contract for computing goods has been established, both the users (through regular payments) and the CC system (by maintaining the requirements) are obligated to follow through with their agreement.