Visit Daniel Farkas to the AMC
Daniel Farkas visited the the e-Bioscience group of the AMC in Amsterdam on March 16-18, 2011 to offer support in de
pilot Desktop Grid project.
This is an activity of the
European Desktop Grid Initiative (EDGI) project.
Daniel Farkas
Daniel Farkas is a researcher at the School of Electronics and Computer Science, University of Westminster, London.
Support on the pilot Desktop Grid project
The main reason for Daniel Farkas to come visit the e-Bioscience group op the AMC is to offer support on the
pilot Desktop Grid project.
Daniel Westerbeek ran into technical issues while carrying out the project and the expertise and experience of Daniel Farkas is much needed to get back on schedule.
The most important thing to do with Daniel Farkas is to set up a working test bed with a working test application. Next, we should try to implement a way to measure calculation time, overall time and use of bandwidth by the application and overhead.
After this has been set up, a demo will be given to Silvia and perhaps Frans Voorbraak, to show that the test bed is really working as it should.
Daniel Westerbeek will also discuss the evaluation strategy for the test bed and pilot desktop grid.
While Daniel Farkas is here, a meeting will be set up to discuss porting a classification application so that they can be deployed on the desktop grid.
Perry Moerland will be involved in this.
To do (all done):
1. Get the test bed working with a sample application, like DSP or Uppercase.
2. Test the test bed with different configurations and for possible errors.
3. Discuss the evaluation plan for the test bed / pilot DG.
4. Program / install measurement tools to carry out the evaluation.
5. Discuss the porting of BWA.
6. Discuss how to deploy a possiblity for users to submit jobs and get results back.
7. Discuss the technical issues Daniel Farkas' team ran into while deploying the DG at the University of Westminster.
8. Give a demo of the working test bed to Silvia and Frans.
9. Discuss the porting of a classification application with Perry.
Full program
March 16, wednesday
| When |
What |
Who |
| - |
Setting up a working test bed. |
Daniel F., Daniel W. |
| - |
Implementing a way to measure calculation time and bandwidth. |
Daniel F., Daniel W. |
March 17, thursday
| When |
What |
Who |
| - |
Discuss evaluation strategy for the DG. |
Daniel F., Daniel W. |
| 3pm - 4pm |
Giving a demo of the testbed to Silvia and Frans Voorbraak. |
Daniel F., Silvia, Frans, Daniel W. |
March 18, friday
| When |
What |
Who |
| 10am - 11am |
Meeting about porting a classification application to the DG environment. |
Daniel F., Perry, Daniel W. |