In the envisioned Internet of Services, commercial users will be able to flexibly and cost-efficiently realize business processes using Web services from external providers. Unfortunately, this also permits attackers to collect potentially sensitive information about the service user’s business activities by simply (and often legally) monitoring the communication endpoints [MSS+11].
In order to alleviate this problem, so-called anonymity networks such as JonDo or Tor can be employed, which obfuscate the relation between two communication parties. However, these mechanisms may inflict the Quality of Service (QoS) properties of the Web services as an undesired side-effect [MLS+10].
WS-Anon.KOM is a comprehensive data set that provides QoS measurements, which have been conducted in a direct fashion and using common anonymity networks, for ten globally distributed Web services. The dataset thereby not only permits the quantification of the QoS-related side-effects of anonymity networks, but also provides an insight into the expected QoS behavior of Web services in the future Internet of Services.
A manuscript that extensively describes both the acquisition and subsequent analysis of this dataset has been accepted at the 2013 Conference on Networked Systems (NetSys 2013). Full bibographical information and a pre-print version of this manuscript are available in the publications section of our lab's Web site [LML+13].
WS-Anon.KOM 1.0 (ZIP archive, 5.6 MB)
WS-Anon contains QoS measurements for the parameters response time, availability, and throughput. The dataset is based on a total of 2.25 million invocations of ten globally distributed Web services. In addition, the archive contains four sample SQL scripts for data analysis.
WS-Anon.KOM 1.0 - Documentation (HTML file, < 10 KB; Revision b)
Provides details about the dataset, including instructions on data import and explanations of the data structure.
For further information or inquiries about WS-Anon.KOM, please contact Ulrich Lampe.