Dear participants!

During ~5 months of life, the project has significantly grown up and we want to thank all participants for that. Today there have been processed more than 1.2 mln of workunits (from ~23.3 mln of the entire set), which gave us several thousands of results with "permutational" orthogonal diagonal Latin squares and interesting search for structures constructed by these squares and squares orthogonal to them (like previously published "Necklace"). We automated many tasks which are needed for everyday project work and consumed most of time, and started deeper work with received results.

Also we encountered mismatch of current format of results publishing (as simple HTML plain page with list of ODLS pairs) with their essence and after publishing last results from December we stop to publish it in this form and start creating another interface for viewing, searching and filtering the results. Of course, we keep the tradition of congratulating participants who found the orthogonal squares. And granting them badges, too. :) And, of course, after completion of postprocessing all results will be published - at least in "raw" format.

Another important thing is the optimization of default app. We plan to implement Daniel's optimizations into it, but after setting up results postprocessing because any participant already can download and use optimized application (for many platforms!) by links from the corresponding thread.

Hope that this "technical update" makes the project life more clear to you. Thank you for your participation and happy crunching!

Minor update at 2018-01-07.

As you might noticed, number of tasks ready to send (line "Tasks ready to send" on the Server status page) became a significant smaller. This is a consequence of new mechanism of tasks generation. It starts up every 5 minutes and if tasks queue is empty then adds a new 32000 tasks in queue.

Minor update at 2018-01-07.

As you might noticed, number of tasks ready to send (line "Tasks ready to send" on the Server status page) became a significant smaller. This is a consequence of new mechanism of tasks generation. It starts up every 5 minutes and if tasks queue is empty then adds a new 32000 tasks in queue.

Please modify this a bit, to add new tasks when there is less that 1000 tasks in the queue. By doing so some tasks should always be in the queue.

---

Hello Daniel!

Please modify this a bit, to add new tasks when there is less that 1000 tasks in the queue. By doing so some tasks should always be in the queue.

What risks are produced with queue exhaustion on time about 2.5 - 3 minutes (in average, by real interval [0;5] minutes)?

As I see, some tasks on long time can fall in queue and not be sent to computers of participants because scheduler choose other, "younger"(!) tasks! And only full exhaustion of queue is need, to force scheduler process this "old" tasks and sent them to computers.

But I not be sure that is a best solution and now a began to monitor of tasks for more precise analysis of them distribution. If I see that "lower border" of unsent tasks is grow continuously, if course I increase the threshold value.

Thank you for attention!

As I see there no problem now and I set the threshold to 1000!

Happy crunching! :)