Sprints 2025 - timetable and projects selected

[UBT - Timbo]

Why is Asteroids@home choosen as Sprint Project?

Hi

The project was chosen (from a range of projects that have plenty of tasks) as the Asteroids admins are doing a good job right now as I have been in direct contact with them about this Sprint. They have even generated more tasks to ensure that FB members can download as many as they might need, over the weekend.

So, give them a break and let's see how it goes.

The admins of Asteroids have demonstrate that they are not able to administrate a Boinc Server handling Sprints in every sprint this Year.

Choosing such a projekt kills the equal opportunity in a Sprint.

Besides, we should not support Admins which kick the supporters of Boinc Projects in the backside region.

Greetings

csbyseti

I would suggest you target the admins of the project with your other comments and opinions.

For me, I see a lot of other projects that have bigger issues than Asteroids. And we all know that many BOINC projects have limited resources, limited funding and they rely on volunteers to help run the projects as well.

Just look here at the number of "Results ready to send" (aka available Work Units) and see which projects are fully active:

https://www.boincstats.com/page/projectStatusInfo

Many stalwarts from years gone by, are now inactive, such as Collatz, Einstein, Rosetta, TN-Grid and even MilkyWay and Moo! Wrapper have very few tasks.

regards
Tim

[UBT - wbiz]

Choosing such a projekt kills the equal opportunity in a Sprint.

I don't see how Asteroids kills equal opportunity, getting work out of Asteroids is very difficult but that is equal for everybody.

I'm running Windows and Linux operating systems, AMD, Intel and ARM processors and they all had their struggles getting work, it doesn't appear biased against any particular type.

To give Asteroids some real credit, I saw their database access crash at about 02:15 on Saturday morning but it recovered almost immediately unlike many other projects that take hours to recover.

The balance between GPU and CPU credits are a lot more balanced than most other projects which I can see might upset those that have invested in powerful GPU cards but that can be seen as fair or unfair.

tldr; it is painful but fair.

[Tamagoch]

Ha all,

the problem should be addressed at Asteroids forum. Despite having millions tasks ready, their feeder seems to be unable to maintain a quick distribution of large batches of tasks. It's happening every day, not only during sprints. Visually I see it in 'Task in progress' number which didn't even change during the sprint. There is some bottleneck for sure.

LG

[UBT - Timbo]

Ha all,

the problem should be addressed at Asteroids forum. Despite having millions tasks ready, their feeder seems to be unable to maintain a quick distribution of large batches of tasks. It's happening every day, not only during sprints. Visually I see it in 'Task in progress' number which didn't even change during the sprint. There is some bottleneck for sure.

LG

Hi

Can I suggest you contact the Asteroids co-admin, "Kyong", via his profile here:

https://asteroidsathome.net/boinc/show_ ... p?userid=1

Or, I can send you a PM with his actual email address as he is one of the admins I am in contact with, regarding Sprints.

Just let him know your "user experiences" regarding the issues you have mentioned, as I am sure he would be glad to hear some feedback and maybe either fix these issues, or give you a explanation as to what processes is happening in the background.

I would guess that part of the problem might be the hardware they are using and how much "CPU power" it has to manage hundreds of simultaneous requests for tasks - a 2019-era 12-core (24 threads) Xeon running at a base speed of 2.2GHz is not exactly a fast processor?:

Supermicro X11SW-TF (1 x Intel Xeon Silver 4214 2.20 GHz, 256 (4 x 64) GB RAM, 8 x 2TB HDD SAS 10K rpm (RAID6) OS Debian Jessie)

regards
Tim

[UBT - Timbo]

Hi all

The 22nd Sprint is starting soon - and the randomly chosen projects has been selected.

PLEASE NOTE the REVISED DATES** and the early start and ending times ! !

From : 2025-11-20 06:00 UTC
To : 2025-11-23 06:00 UTC

** Yes, I got my dates mixed up, due to it being a NIGHT RACE - but the website has been updated now..

regards
Tim

[UBT - Timbo]

Hi all

The projects for Sprint #23 AND Sprint #24 have now been randomly selected.

UPDATE: Sprint #23 is for NumberFields

Good luck to everyone and I hope everything works smoothly, as the admins for both projects will do their best to ensure there are sufficient tasks available for all FB members to crunch.

regards
Tim

[stream]

Just look here at the number of "Results ready to send" (aka available Work Units) and see which projects are fully active:

https://www.boincstats.com/page/projectStatusInfo

Be careful with this page, it's partially broken.
It says, for example, that Einstein and Private GFN has no task available, but it's not true. These projects are correctly monitored by other site:
Einstein
GFN

[yoyo]

Is there a list of projects, which can be taken as Sprint Projects?

I have the feeling, that your random project Sprint selection doesn't work correct. Some projects are already 3 times Sprint project and some projects not even once. E.g. the projects yoyo@home and yafu have always workunits available and are not choosen as Sprint project?

[UBT - Timbo]

Is there a list of projects, which can be taken as Sprint Projects?

Hi

Yes there is a list of Sprint projects (for 2025) here:

viewtopic.php?t=412

I have the feeling, that your random project Sprint selection doesn't work correct. Some projects are already 3 times Sprint project and some projects not even once. E.g. the projects yoyo@home and yafu have always workunits available and are not choosen as Sprint project?

The Project selection process is quite simple:
1) The project must plan to be online during the relevant dates
2) The project must have had significant numbers of tasks available for a week or two before the start date, which is monitored by me
3) The project must show that they can support the much higher demand for tasks in terms of bandwidth, and server response times
4) An attempt is made to contact the project admin(s) in advance to check that they can cope with the demand and that they are not planning any maintenance during the weekend - sometimes admins do NOT respond so they cannot go into the draw to be selected.
5) The same project cannot be selected in consecutive Sprints

Hence from the "selectable" list, so projects are immediately discounted from selection, due to any of the above reasons. eg: GPUGrid has no tasks; WCG has limited numbers of tasks available AND they do not answer emails; Einstein, Moo! Wrapper, Rosetta, SiDock and TN-Grid have consistently had limited numbers of tasks available; Asteroids and Gerasim have had operational difficulties; and my old notes (from maybe 2 years ago) mention that the Private GFN Server admin requested to be omitted from Sprints.

Unfortunately, some projects rule themselves out of selection due to at least one of the above "rules" being "true". So, this then limits the number of projects that go into the "hat" and hence, for some Sprints, the choice is very limited and hence some repetition is to be expected during the course of the 24 Sprints held each year.

regards
Tim

[yoyo]

That keeps the question open why yoyo@home wasn't selected even once and some projects are already the 3rd time a Sprint project.
Which rule has yoyo@home not fullfilled? There are always a lot of jobs available.