The work generator is stopped. BOINC server will wait for all WUs that are in-progress. The site and the forum will continue to work during the maintenance. However, the registration of new users will be suspended. The TODO list can be found here.

The alpha-testing was successful! Here are some results and conclusions.

• About 5.6% (~20,500 structures) of COD entries are processed. The collected data will be used to determine the optimal set of crystallite sizes and the optimal resolution of the simulated diffraction patterns for the final database.
  
• A few critical bugs (see e.g., 1, 2 and 3) in the app are fixed thanks to volunteers who participated in debugging on this site and on http://boinc.ru. All apps are sufficiently stable now. A fraction of fails is about 1% (see, e.g., these statistics for November 27th).
  
• The site functionality was improved thanks to the advices given by volunteers.
  
• The optimal parameters of the virtual server are determined: 2 cores of Intel Xeon E5-2666, 4 GB RAM, 500 GB SSD storage with read/wright throughput of 1500 IOPS, network load: 60 MB/s.
  
• The processing power donated by volunteers to this project was much higher than expected. This allows to extend the initial range of crystallite sizes to larger sizes (up to 30 nm).
  
• Although the MultiSize feature has slightly improved the scheduling, it is still far from optimal. It is clear now that CPU and GPU apps should receive WUs of different size due to large difference in performance. It is also clear that single-threaded CPU app is required to process small WUs. So, the distribution of WUs for different crystallite sizes between the apps should be like this:
  
  CPU, single-threaded: < 10 nm
  CPU, multi-threaded: 10 – 20 nm
  GPU: > 20 nm

I will post the progress of work from time to time. However, please do not expect much until the spring of 2017.

I am very grateful to all volunteers participating in the alpha-testing! I apologize for poor server performance, awful job scheduling, app crashes and bugs leading to a waste of computational resources. Thank you for your patience and understanding!