Cloud Armor develops a cloud service crawler engine that collected cloud services available on the Web. After parsing more than half million possible links, Cloud Armor manages to identify and collect meta-data for closely 6,000 real cloud services. From the collected information, we prepared several large datasets of real-world cloud services.

Main Cloud Services Dataset

We explored the Web for cloud services and the Cloud Armor managed to parse 617,285 possible links and discover 34,348 possible seeds for cloud services. The discovered cloud service Web pages are full of noisy data (e.g., wiki, news, articles that are related to cloud services) which need to be filtered. After the filtration process, Cloud Armor identified 6,686 unique cloud services. It should be noted that these cloud services were collected late 2012 and some might not be available anymore. We will perform another round of data collection late 2013. The links of the cloud services collected in 2012 can be downloaded from here [67K].

WSDL & WADL Dataset

The crawler engine searches cloud services implemented using WSDL and WADL. Our Cloud Armor collects 1,539 possible links for cloud services in WSDL and WADL. After the filtration process, Cloud Armor identifies 118 valid cloud services implemented using WSDL and WADL. The collection can be donwloaded from here [806K].

Trust Feedback Dataset

In order to bootstrap Cloud Armor trust management service, we collected cloud service consumers' feedback from leading review websites such as Cloud Hosting Reviews, Best Cloud Computing Providers and Cloud Storage Reviews and Ratings. Cloud Armor collects 10,000+ feedbacks given by nearly 7,000 consumers to 113 real-world cloud services. The feedbacks are based on Quality of Service (QoS) attributes. The dataset can be downloaded from here [746K]. Note that the information of consumers was anonymized.