Lone, Q., Luckie, M., Korczynski, M., Asghari, H., Javed, M., & van Eeten, M. (2018). Using crowdsourcing marketplaces for network measurements: the case of Spoofer. In Proceeding of 2nd Network Traffic Measurement and Analysis Conference (TMA 2018) (pp. 1–8). Washington DC, USA: IEEE. https://doi.org/10.23919/TMA.2018.8506499
Permanent Research Commons link: https://hdl.handle.net/10289/12500
Internet measurement tools are used to make inferences about network policies and practices across the Internet, such as censorship, traffic manipulation, bandwidth, and security measures. Some tools must be run from vantage points within individual networks, so are dependent on volunteer recruitment. A small pool of volunteers limits the impact of these tools. Crowdsourcing marketplaces can potentially recruit workers to run tools from networks not covered by the volunteer pool. We design an infrastructure to collect and synchronize measurements from five crowdsourcing platforms, and use that infrastructure to collect data on network source address validation policies for CAIDA's Spoofer project. In six weeks we increased the coverage of Spoofer measurements by recruiting 1519 workers from within 91 countries and 784 unique ASes for 2,000 Euro; 342 of these ASes were not previously covered, and represent a 15% increase in ASes over the prior 12 months. We describe lessons learned in recruiting and renumerating workers; in particular, strategies to address worker behavior when workers are screened because of overlap in the volunteer pool.
This is an author’s accepted version of an article published in Proceeding of 2nd Network Traffic Measurement and Analysis Conference (TMA 2018). ©2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.