主持人｜賈尼瑪
Internet of Things (IoT) is anticipated to bridge various technologies to allow new applications through linking physical things together in the future. Cloud computing is becoming a prevalent business paradigm for software delivery and services, allowing businesses to save on the costs of infrastructure, maintenance, and personnel. IoT applications aim to provide access to widespread interconnected networks of smart devices, services, and information.Although the cloud services provide virtually unlimited resources, they are limited in terms of scope. Therefore, the cloud integrated IoT, has provided an unprecedented opportunity for developers to develop more value-added software services. Services in cloud computing can be categorized into application services and computing services. Increasing facilities through the composition of services are considered as an essential module in this technology. Several types of services can be supplied in the IoT using the user’s requirements to consider the user needs. The proper existing atomic services finding to create the required composite service with their cooperation in an orchestration model, regarding their Quality of Service (QoS) properties is species as an NP-hard problem. In this paper, we propose a service composition method using Grey Wolf Optimization (GWO) and MapReduce framework to finding composed services with optimized QoS. The simulation results show our proposed method improve cost, availability, response time and energy saving.
The course of development with steps
The cloud services technology creates the opportunity for building composite services by combining existing elementary or complex services from different enterprises and in turn offering them as high-level services or processes. With more cloud services available, Quality of Service (QoS) becomes an important selling and differentiating point of cloud services. QoS analysis becomes increasingly challenging and important when complex and mission-critical applications are built on services with different QoS. Knowing the quality of a service, particularly of a composite service, is of significance both to the provider and to the client. A service provider needs to know if the quality of the delivered service meets the requirements of the client when the service is composed of several services in a particular way and if the right component services are selected. Also, each cloud service has some QoS factors for service evaluating. The mentioned factors will be evaluated in the total possible compounds of cloud services. QoS-based cloud service selection is a critical process that directly determines the QoS values of the composite cloud service. In this paper, five major QoS attributes are considered: cost, response time, availability, efficiency, and optimization.
Cost
Response time
Availability
Energy Saving
The QoS attributes of a service evaluated during design and execution time. At design time, these attributes help to build a composite service based on the QoS requirements of the user; while at execution time, they can be monitored to maintain the desired QoS level. Information about these attributes can be obtained from the service’s profile, nevertheless, when this information is not available, it can be obtained by analyzing data collected from previous invocations. In the following, we present existing QoS metrics based on the normalization method to recognize the final candidate service composition according to the aggregation of QoS metrics.
A composite service is specified as a collection of abstract application services according to a combination of control-flow and data-flow. Similar to traditional service composition, cloud service composition is conducted in two steps. First, a composition schema is constructed for a composition request. Second, the optimal composition plan is selected. Control-flow graphs are represented using Unified Modeling Language (UML) activity diagrams. Each node in the graph is an abstract application service. There are four control-flow patterns for defining a composite cloud service (composition schema), such as sequential, parallel, conditional and iterative (loop) patterns. Any solution to a composition problem in cloud computing includes: (1) Map the abstract application services to concrete application services and corresponding computing service (VM, database, and network services). (2) Schedule the execution order of the application services. In this research, we introduce only the selection of the composition plan (abstract application services) based solely on user’s QoS attributes.
Results and derivative value
Reducing Energy Consumption
The proposed method is compared to the evolutionary algorithms of GA, ACO, and GWO. The proposed method has been able to reduce energy consumption.
Reducing response time
The proposed method is compared to the evolutionary algorithms of GA, ACO, and GWO. The proposed method decreased the response time compared to the GA and GWO algorithms. However, the performance of the ACO algorithm was better than the method proposed in this paper with these experimental data.
Increasing availability
The proposed method is compared to the evolutionary algorithms of GA, ACO, and GWO. The performance of the GA is better than the proposed method. However, the proposed method was able to increase availability better than the other two methods.
Reducing Cost
The proposed method is compared to the evolutionary algorithms of GA, ACO, and GWO. The proposed method was able to reduce the cost compared to other algorithms.
No direction
In service composition QoS some parameters are existing as scalability, efficiency, optimization, privacy and etc. Improving these parameters are our future research directions. Moreover, one of important problem in cloud-integration IoT service composition is user’s privacy, in future work, we focused on increasing privacy.
Comments