Research Use of Seattle

We have used Seattle to construct a network testbed. The purpose of the Seattle testbed is to support cutting-edge research over the Internet through the use of real, end-user machines. Our first and foremost goal is the safety of the participating machines, so there are some restrictions on what one can do on our testbed. For example, programs cannot generate ICMP traffic, modify the host kernel, etc. (other excellent testbeds, like Emulab and PlanetLab, provide low-level access to end hosts). Our focus is to attract real Internet users to participate in our testbed. This will increase the heterogeneity, scale, and diversity beyond what is achievable through other means.

Quick Links


Q: What are the restrictions on how I use Seattle?
A: See our Acceptable Use Policy.

Q: What is Seattle Clearinghouse and how does it differ from Seattle?
A: Seattle is the core software that all end-machines run. Seattle Clearinghouse is a communal website that allows users to pool and share Seattle resources. Resources that are allocated to Seattle Clearinghouse result in the researcher ceding direct control over the donated resources.

Q: What if I want to control resources from my installer directly?
A: You can also build an installer that allows you direct control of resources. Keep in mind, you won't get credited through the Seattle Clearinghouse website, etc.

Q: I want to run an experiment on more vessels than I am currently granted. How can I do this?
A: You can have more people install Seattle on your behalf. You can download the installers that will credit new user resources to your account from within Seattle Clearinghouse. You gain access to 10 vessels for each new install.

Q: I want a vessel with more resources than you allow. How can I get this?
A: You can have people perform installations you directly control with donations that exceed 10% of their machine resources. This will grant you vessels that have more resources than the default.

Q: What sorts of machines are on the Seattle testbed?
A: The Seattle testbed includes:

Q: How do I get machines of type X?
A: There is an external group that is working on a mash-up on the Seattle Clearinghouse sharing site that will provide detailed node information. You're also free to monitor the node characteristics you care about (latency, loss, availability, etc.) and provide these publicly. We'd be happy to point other users to your data.

Q: I want to run C code / use the GPU / send traffic with spoofed IP addresses. How do I?
A: We don't support these sorts of things for safety reasons. Our primary concern is the safety of our end users.

Q: My question is not answered. Who do I contact?
A: Please contact our team at if you have any further questions concerning Seattle.

This material is based upon work supported by the National Science Foundation under Grant Numbers (0834243, 1205415, and 1223588). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.