CNMM, Tsinghua University, Beijing, China and Citizen Cyberscience Centre, Geneva, Switzerland
25 Apr 2014
Summary Researchers announce the end of the project as they complete further studies to confirm interesting results gained from World Community Grid data and prepare to publish their exciting findings.
Over the past six months, we have been doing further analysis of the results we obtained from the Computing for Clean Water project, in order to finalize a manuscript for publication. We've now reached a level of confidence in our findings that we don't believe any further runs on World Community Grid are necessary, which marks the end of the project on World Community Grid. We aim to submit our manuscript in the coming few weeks, once all co-authors have had a chance to review it.
As we've mentioned before, the results we got from Computing for Clean Water were surprising in several ways. Since Computing for Clean Water allows us to effectively extend simulations to a new flow regime that is closer to what happens in real experiments (see the diagram below), we're being very cautious about testing our surprising results in different ways, to make sure that what we see is a genuine effect, and not some unexpected artifact of the way we are doing the calculations.
We'll be able to tell you more about exactly why our results are so surprising once the article has been submitted for publication. At that point, we'll release a so-called electronic preprint, so that colleagues – and you, the Computing for Clean Water volunteers – can see the results, too. For the moment, though, suffice to say that we're very excited about having discovered a new mechanism that could make water filtration by nanotubes much more efficient. This mechanism appears to have been overlooked in previous studies because they did not have the computing power to simulate the flow process in the sort of detail that we can with Computing for Clean Water.
The diagram below succinctly captures the big step forward that Computing for Clean Water enabled us to make in terms of bridging the gap between the (high) flow velocity that prior simulations have managed to study, and the (low) range of flow velocities for which practical experiments on water flow in nanotubes have been carried out.
So you may be wondering, what exactly have we been doing for the last few months, since we stopped running jobs on Computing for Clean Water? Well, the first step, before asking for independent feedback from reviewers, is to have scientists involved in the project directly ask tough questions about the results. To answer those questions we have needed to do further computations, carried out on computing clusters in the UK and in Australia. (Technical aside: these tests represent a much smaller volume of calculations than the Computing for Clean Water project. However, each calculation generates much larger data files than a corresponding Computing for Clean Water task. So doing these tests on a dedicated computing resource makes sense.)
These further computations supported the initial conclusions we drew from analyzing the Computing for Clean Water data. And as mentioned above, this gives us the confidence we needed to move ahead with publishing these findings.
We want to sincerely thank World Community Grid volunteers for supporting our work and for allowing us to extend our simulations to such a realistic level. While our work on World Community Grid comes to an end, our research efforts continue and we look forward to sharing details of our exciting findings and any further developments in the coming months.