For the “CERN Public Computing Challenge 2015”, CERN is asking volunteers to contribute their spare computing power to help CERN scientists simulate billions of particle collisions, in order to compare theoretical models with experimental results from CERN’s Large Hadron Collider and other particle colliders.
In particular, this one-month challenge is a chance for scientists to test new technologies for distributing and running simulations for the LHC experiments, using publicly contributed computing power. Namely, we’ll be testing new ways of scheduling the simulations we’re sending out, using a CERN technology called Data Bridge, and a new, lightweight version of the CernVM technology which runs the simulations.
For this year’s challenge, CERN is also asking for your help with new communities and social networks, in order to achieve a better gender balance as well as greater linguistic diversity amongst those contributing to the challenge. To help with this, we’re testing a new credit and badging system called CreditPiggy, which we hope will motivate your participation with both individual and challenge-wide statistics about your contributions.
Below you can see the CreditPiggy goals for the whole challenge:
We’ll be sharing our progress towards these goals throughout the challenge, together with details about your own contribution. And if you have any questions, about the science, the software or the social aspects of this Challenge, don’t hesitate to post them in the forum at the bottom of this page.
The first CERN Public Computing Challenge ran for 12 days in December 2014, as part of CERN’s 60th anniversary celebrations. We invited volunteers around the world to help CERN scientists simulate particle collisions in accelerators like the Large Hadron Collider (LHC), using their own computers. The challenge was an opportunity for us to test new, simpler approaches to distributing such computations, with the help of CERN’s own virtual machine technology, CernVM.
Thanks to efforts of the volunteers, we were able to simulate over 19 billion particle collisions. The results are being used in a new educational game called Virtual Atom Smasher. And thanks to detailed feedback from volunteers, we were able to make big improvements in the software used to manage the challenge.
The first time you join the challenge, you will be prompted to install an application called “CernVM WebAPI”. This takes care of delivering the required software for the challenge to your machine.
What is this, really?
This is a tiny system service (about 1MB) which provides the interface between the web browser and a Virtual Machine in your computer.
We have packed all the scientific software required to run the challenge into a compact, virtual computer. Virtualbox is a free, open-source software that runs this virtual computer.
Why is this necessary?
The scientific software that we use runs only on Linux. That’s why we packed our Linux distribution in a virtual machine so it can run on almost any operating system.
After all this software is installed on your computer, just click “Start” on the Challenge Dashboard, wait for the software to load and join the challenge! You can press “Stop” at any time. Note that closing the Challenge Dashboard does not stop the Virtual Machine!
What can I control?
You can define how much CPU and RAM you allocate to the project, by clicking on the gear icon next to the start button. Remember to click Apply!
killall VBoxHeadlessand hit enter.
”C:\Program Files (x86)\CERN\CernVMWebAPI\cernvm-webapi.exe" service
launchctl unload ~/Library/LaunchAgents/cern.cernvm.webapi.service.plist
sudo apt-get remove cernvm-webapi
sudo yum remove cernvm-webapi
We are using Google Analytics on the Computing Challenge and Web API pages, in order to gather detailed statistics about the website's traffic and the system's behavior. This information allows us to evaluate our technology, and continue improving it. In particular, we are curious to discover where people may get stuck with the instructions, and what other projects they may click through to, from this site.
This data is being gathered for research purposes in the Citizen Cyberlab project, where we’re trying to understand what people learn about science from participating in such initiatives, and how we might improve the learning experience in future. According to the Google Analytics terms of service, we do not track, collect or upload any data that personally identifies an individual (such as name or email).