Cloud computing makes our life easy by delivering computing resources as utility like telephony, water and gas. In cloud computing users should pay only for what they consumed. Nowadays, cloud service providers deliver a huge number of cloud services with almost the same features which makes the cloud services discovery and selection process as a big challenge for the end consumers. Using existing search engines results in a lot of unrelated outcome which increases the cloud service discovery and selection process time and effort. In this paper, we present an enhanced cloud services marketplace framework to facility the cloud services trading between providers and consumers and to make could services more visible for all consumers. Proposed framework receives users' requests as a voice commands or flat-text then translates them based on Natural Language Understanding technologies. In additional, we enhanced the matching algorithm by adding different weights for attributes based on consumer preferences. Experiments showed an enhancement in the overall user experience and better matching for user request. Keywords: Cloud computing, cloud services, service matching, cloud marketplace, cloud architecture.
I. INTRODUCTIONCloud computing [1] is a paradigm that enables access to a shared computing resources where consumer should pay only for what he consumed. There are three different levels of cloud services. Infrastructure as a Service (IaaS) where users hire an infrastructure resources from another party , Platform as a Services (PaaS) where users develop applications on a haired platform and Software as a Service where users use an applications developed and operated by another party based on pay as you use model. Nowadays, there are big number of cloud service providers who provide a huge number of similar cloud services which makes the process of finding and selecting the appropriate cloud service a tedious task for the users. Using general search engines like Google increases the time and effort of finding the cloud services. In this paper, we present an enhanced architecture for cloud services marketplace based on the work proposed on [2]. Enhanced architecture allows users to enter the cloud service requests as a voice commands or as plain texts. In addition, enhanced architecture improved the matching algorithm by applying different weights for user requested attributes based on user preferences. . Experiments showed an enhancement in the overall user experience and better matching for user requests. The rest of this paper is organised as following: section 2 surveys the related work, section 3 presents the enhancement architecture, section four displays the results of the experiments and section 5 is a conclusion and future work.