Memory usage for GC1 is up again, I see. Now using 198,636KB of RAM and 197,876KB of VM on one of my computers. It had crashed a couple of the tasks as Windows had run out of VM and was increasing the page file... not weird when the app is eating it like cookies.
Memory usage for GC1 is up again, I see. Now using 198,636KB of RAM and 197,876KB of VM on one of my computers. It had crashed a couple of the tasks as Windows had run out of VM and was increasing the page file... not weird when the app is eating it like cookies.
My big guy certainly performs better with an upgrade of RAM ( to 10G ) and VM ( disk of 1TB ) to satisfy 8 cores worth of the munchies. Yup, it's a cookie fest.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
So, the project now is working fine and it seems that we can forget about it for about a year while the current search is runnig. The power of our computers grows in orders of magnitude, but the volume of computations grows in the same manner, while we do not see any obvious advantage of this in science field. And it feels sad if nothing new will happen in the near future. Are there any news about optimized apps, that will help us finish this run till the end of summer?
So, the project now is working fine and it seems that we can forget about it for about a year while the current search is runnig. The power of our computers grows in orders of magnitude, but the volume of computations grows in the same manner, while we do not see any obvious advantage of this in science field. And it feels sad if nothing new will happen in the near future. Are there any news about optimized apps, that will help us finish this run till the end of summer?
I'd guess that the current new GC code still has potential for some optimization. No actual news yet, but I'll continue to work on this during the run. There are still some things to do behind the scenes first, such that the server infrastructure could better handle the load of faster Apps and shorter running tasks.
One should also not forget that there has been a noticable optimization only recently (in the transition from S5GCE to S5GC1) where the code performance was improved by a factor of 2 to 3, so that more science could be packaged into a single WU (to get about the same overall WU runtime). So it's not like there is a standstill.
So, the project now is working fine and it seems that we can forget about it for about a year while the current search is runnig. The power of our computers grows in orders of magnitude, but the volume of computations grows in the same manner, while we do not see any obvious advantage of this in science field. And it feels sad if nothing new will happen in the near future. Are there any news about optimized apps, that will help us finish this run till the end of summer?
I'd guess that the current new GC code still has potential for some optimization. No actual news yet, but I'll continue to work on this during the run. There are still some things to do behind the scenes first, such that the server infrastructure could better handle the load of faster Apps and shorter running tasks.
Please, keep us in touch as often as you can, so that we'll not feel the project is working independent of us. It will be great to publish a technical diary (or blog, like Technical news in SETI) with daily or may be weekly reports.
Please, keep us in touch as often as you can, so that we'll not feel the project is working independent of us. It will be great to publish a technical diary (or blog, like Technical news in SETI) with daily or may be weekly reports.
Thanks in advance.
That sounds like a good idea. Daily updates surely won't work, but roughly weekly should be feasible. Note that none of the Devs is working for E@H exclusively, there are days and even weeks when we don't get to anything related to E@H.
We're still working on the (overdue) web code update, I'm thinking of adding a "technical news" forum (Devs post only) in the new code. How does SETI do it?
We're still working on the (overdue) web code update, I'm thinking of adding a "technical news" forum (Devs post only) in the new code. How does SETI do it?
Same way as the new News forum is done.
See in one of the newer forum.inc files:
function user_can_create_thread($user, $forum) {
if (!$user) return false;
if ($forum->is_dev_blog){
if (
(!$user->prefs->privilege(S_SCIENTIST)) &&
(!$user->prefs->privilege(S_DEV)) &&
(!$user->prefs->privilege(S_ADMIN))
) {
return false;
}
}
return true;
}
See [trac]changeset:19949[/trac] and Project News.
Quote:
News items are stored as threads in a special forum, named 'News' by default (you can change this by defining a constant NEWS_FORUM_NAME in project.inc). This forum should have 'is_dev_blog' set to 1, so that only project admins will be able to create threads (non-admins will, however, be able to reply to threads). The forum-creation script ops/create_forums.php creates this forum.
BTW, why there are so many results in database for ABP2 and S5GC1 tasks and no WUs with no final result for S5GC1? WUs queue was increased by the order of mangnitude or something else?
Memory usage for GC1 is up
)
Memory usage for GC1 is up again, I see. Now using 198,636KB of RAM and 197,876KB of VM on one of my computers. It had crashed a couple of the tasks as Windows had run out of VM and was increasing the page file... not weird when the app is eating it like cookies.
RE: Memory usage for GC1 is
)
My big guy certainly performs better with an upgrade of RAM ( to 10G ) and VM ( disk of 1TB ) to satisfy 8 cores worth of the munchies. Yup, it's a cookie fest.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
So, the project now is
)
So, the project now is working fine and it seems that we can forget about it for about a year while the current search is runnig. The power of our computers grows in orders of magnitude, but the volume of computations grows in the same manner, while we do not see any obvious advantage of this in science field. And it feels sad if nothing new will happen in the near future. Are there any news about optimized apps, that will help us finish this run till the end of summer?
RE: So, the project now is
)
I'd guess that the current new GC code still has potential for some optimization. No actual news yet, but I'll continue to work on this during the run. There are still some things to do behind the scenes first, such that the server infrastructure could better handle the load of faster Apps and shorter running tasks.
BM
BM
One should also not forget
)
One should also not forget that there has been a noticable optimization only recently (in the transition from S5GCE to S5GC1) where the code performance was improved by a factor of 2 to 3, so that more science could be packaged into a single WU (to get about the same overall WU runtime). So it's not like there is a standstill.
CU
HB
RE: faster Apps and shorter
)
With less memory use, I hope?
RE: RE: So, the project
)
Please, keep us in touch as often as you can, so that we'll not feel the project is working independent of us. It will be great to publish a technical diary (or blog, like Technical news in SETI) with daily or may be weekly reports.
Thanks in advance.
RE: Please, keep us in
)
That sounds like a good idea. Daily updates surely won't work, but roughly weekly should be feasible. Note that none of the Devs is working for E@H exclusively, there are days and even weeks when we don't get to anything related to E@H.
We're still working on the (overdue) web code update, I'm thinking of adding a "technical news" forum (Devs post only) in the new code. How does SETI do it?
BM
BM
RE: We're still working on
)
Same way as the new News forum is done.
See in one of the newer forum.inc files:
function user_can_create_thread($user, $forum) { if (!$user) return false; if ($forum->is_dev_blog){ if ( (!$user->prefs->privilege(S_SCIENTIST)) && (!$user->prefs->privilege(S_DEV)) && (!$user->prefs->privilege(S_ADMIN)) ) { return false; } } return true; }See [trac]changeset:19949[/trac] and Project News.
BTW, why there are so many
)
BTW, why there are so many results in database for ABP2 and S5GC1 tasks and no WUs with no final result for S5GC1? WUs queue was increased by the order of mangnitude or something else?