News General

No, it is set to run on preferences, and all other projects run at configured times. BOINC says it is a MilkyWay bug. Is there a better place to report bugs?

Still not fixed, still running when forbidden, so I must manually suspend all BOINC processing to stifle it (forgot to turn it back on yesterday after tutoring a student):

https://drive.google.com/file/d/103iodHcPHN_MoEJz51gPmuWO2ZaQBpwx/view?usp=sharing

BTW, the lack of suspension is not a solid defect. Sometimes it suspends as it should. Perhaps it does not check during a prolonged loop.

BTW, the lack of suspension is not a solid defect. Sometimes it suspends as it should. Perhaps it does not check during a prolonged loop.

Are your tasks in danger of expiring and could that be why Boinc isn't suspending them?

BOINC does not do the suspending, the app does, and the implicit contract is that if it should suspend, it must, even if it misses a deadline from the gods of the app. And eventually or sometimes it does, hence the long loop theory. One MilkyWay task is suspended right now.

Olá, gostaria de melhor aproveitar minha placa de vídeo. Antes eu usava para minerar bitcoins, acredito ter uma potencia razoavel para contribuir em Milkway.

Quando encontrarem algum upgrade do software por favor me avisem.

Hey everyone,

I've been traveling a bunch this summer for conferences (and for fun), and I've also been busy with some new projects for my thesis. Apologies for not paying too much attention to the forums over this time! I've been keeping an eye on the server though, which luckily hasn't exploded.

A few things:
(1) We have two new scientists, Kevin Roux and Hiroka Warren, who are PhD students working on the N-body project! As I'm transitioning away from working on milkyway over the next year, they'll be stepping in to fill my shoes and transition the project to its next stages.
(2) Thanks to several generous donations from milkyway volunteers, we have just bought a new server machine with 8 TB of storage in RAID 1. This means that in the case of a hard drive failure, the server shouldn't slow down at all, and we won't have a repeat of this year's server disaster. Also, the new drives are SSDs, not HDDs, so hopefully we should notice a speed increase. We will be working on migrating to this new server hardware, and we would like to have this done before the holidays.

@ Tom, how many simultaneous connections are you seeing these days?

Something like ~30k computers are running at any given time, which is only ~10-20 server connections at any given moment, based on what I see when I'm staring at the DB processes.

Thanks for your continued support!

Best,
Tom

Oh, and you're welcome to say hi to Kevin and Hiroka at these news posts: https://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=4928 and https://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=4927.

One thing that they're working on is updating the publications page. Hopefully soon you should see some progress on that!

Hey everyone,

I've been traveling a bunch this summer for conferences (and for fun), and I've also been busy with some new projects for my thesis. Apologies for not paying too much attention to the forums over this time! I've been keeping an eye on the server though, which luckily hasn't exploded.

A few things:
(1) We have two new scientists, Kevin Roux and Hiroka Warren, who are PhD students working on the N-body project! As I'm transitioning away from working on milkyway over the next year, they'll be stepping in to fill my shoes and transition the project to its next stages.
(2) Thanks to several generous donations from milkyway volunteers, we have just bought a new server machine with 8 TB of storage in RAID 1. This means that in the case of a hard drive failure, the server shouldn't slow down at all, and we won't have a repeat of this year's server disaster. Also, the new drives are SSDs, not HDDs, so hopefully we should notice a speed increase. We will be working on migrating to this new server hardware, and we would like to have this done before the holidays.

@ Tom, how many simultaneous connections are you seeing these days?

Something like ~30k computers are running at any given time, which is only ~10-20 server connections at any given moment, based on what I see when I'm staring at the DB processes.

Thanks for your continued support!

Best,
Tom

This is excellent news all around! Good deal!

What about recompiling server?

...
(2) Thanks to several generous donations from milkyway volunteers, we have just bought a new server machine with 8 TB of storage in RAID 1 ...
Tom

Well, finally we see some progress.
Thanks for your work!
cheers

What about recompiling server?

The binaries will all need to be recompiled on the new server hardware. I have been keeping a list of some of the changes people have requested to our server software, but if you have suggestions feel free to share them here! When we recompile, it makes sense to do things like change the task pool sizes, remove the sleep loop in the transitioner, etc.

What about recompiling server?

The binaries will all need to be recompiled on the new server hardware. I have been keeping a list of some of the changes people have requested to our server software, but if you have suggestions feel free to share them here! When we recompile, it makes sense to do things like change the task pool sizes, remove the sleep loop in the transitioner, etc.

I think an option to get "wingman only" tasks would be nice. Hopefully that can be done by some settings at compile time.

Separate selection for "Milkyway@home Separation cpu" and "Milkyway@home Separation gpu" probably also separate selection for "Milkyway@home N-Body Simulation cpu" and "Milkyway@home N-Body Simulation gpu" in the future

Separate selection for "Milkyway@home Separation cpu" and "Milkyway@home Separation gpu" probably also separate selection for "Milkyway@home N-Body Simulation cpu" and "Milkyway@home N-Body Simulation gpu" in the future

It would be VERY nice if we could have `separation` of the work unit types to select .

What about recompiling server?

The binaries will all need to be recompiled on the new server hardware. I have been keeping a list of some of the changes people have requested to our server software, but if you have suggestions feel free to share them here! When we recompile, it makes sense to do things like change the task pool sizes, remove the sleep loop in the transitioner, etc.

Hell, yes. This please!

---

What about recompiling server?

The binaries will all need to be recompiled on the new server hardware. I have been keeping a list of some of the changes people have requested to our server software, but if you have suggestions feel free to share them here! When we recompile, it makes sense to do things like change the task pool sizes, remove the sleep loop in the transitioner, etc.

Hell, yes. This please!

I thought the standard transitioner only slept if it tries to do a pass and can't find anything to do! (I'm willing to be proven wrong...)

Cheers - Al.

Just restarted the processes and flushed out some stale tasks from the DB. Hopefully that fixes whatever was causing the server to load slowly. It seems to be running more quickly for me already.

Separate selection for "Milkyway@home Separation cpu" and "Milkyway@home Separation gpu"

Help !
I update my mac to macOS 13.3.1 and nothing works anymore, for milky way at home... anybody else has the same problem ??? Marc, Switzerland.