Does larger cache or disks with ssd buffers help in this case?
We copied over about 1.6TB of "old" result files (uploaded before the crash); I guess about 500-800 GB were uploaded afterwards, so a RAM cache would need to have at least 500 GB to make a difference.
We don't have a stack of 12 or 16 spare hybrid HDs lying around here, so these would need to be chosen, ordered, delivered and built in under full load (two at once, then wait for the RAID rebuild).
As the alternative is to just wait a few days for finishing the backlogs, neither of the above options I would consider an efficient use of our working time and the taxpayer's money.
As far as our fellow participants and their BOINC clients are concerned, both EInstein@Home and Albert@Home are now working as before. Behind the scenes there is still a lot of work to do, though, but this is less urgent.
Perhaps, whilst you weren't looking, of course, the Tooth Fairy maliciously interfered with your BOINC Manager to cancel NNT for Einstein and a whole flood of new tasks arrived all at once (145 to be precise) :-).
That's gonna keep you busy for quite a while ;-).
Better get cracking then ... :-).
Oh, and it might be a good idea to put your machine in a Tooth Fairy proof enclosure whilst you're at it ;-).
The problem with the upload server was resolved some time ago and as far as I'm aware, all outstanding uploads were cleared soon afterwards. I don't know of any recurrence of a problem with uploading completed results.
When you say "past task completions" are you referring to tasks completed before the upload server was brought back on line - ie prior to 14th Jan?
Can you tell us what particular science run these tasks are from, when they were completed, and what messages you get in the BOINC event log when you try to upload them?
This seemed so strange that I just had a look through your host's task list and you only have 8 'in-progress' tasks, the oldest of which was sent to you around 24 hours ago. It's a bit hard to imagine you have any outstanding uploads. Perhaps you're referring to something else other than uploads so maybe you could rephrase your question?
Anynoe know if there are other Boinc projects thar utilise the screen saver
function ? Seti And Einstein are already running. I would like to add a few more.[/i]
RE: RE: validation is
)
We copied over about 1.6TB of "old" result files (uploaded before the crash); I guess about 500-800 GB were uploaded afterwards, so a RAM cache would need to have at least 500 GB to make a difference.
We don't have a stack of 12 or 16 spare hybrid HDs lying around here, so these would need to be chosen, ordered, delivered and built in under full load (two at once, then wait for the RAID rebuild).
As the alternative is to just wait a few days for finishing the backlogs, neither of the above options I would consider an efficient use of our working time and the taxpayer's money.
BM
BM
As a side-note: Albert@Home
)
As a side-note: Albert@Home is also online again...
Oliver
Einstein@Home Project
As far as our fellow
)
As far as our fellow participants and their BOINC clients are concerned, both EInstein@Home and Albert@Home are now working as before. Behind the scenes there is still a lot of work to do, though, but this is less urgent.
Have a nice weekend, everyone!
BM
BM
RE: Have a nice weekend,
)
And the same to you, Bernd (and Oliver) - you need it and deserve it.
At some future date when your workload allows (but not now), could we remind ourselves to take a look at the intel_gpu OpenCL driver issues, please?
RE: At some future date
)
Yep. This is why I'm not yet deleting BRP4 files.
BM
BM
I hadn't realized that I'm
)
I hadn't realized that I'm working on Einstein WUs. :-)
Perhaps, whilst you weren't
)
Perhaps, whilst you weren't looking, of course, the Tooth Fairy maliciously interfered with your BOINC Manager to cancel NNT for Einstein and a whole flood of new tasks arrived all at once (145 to be precise) :-).
That's gonna keep you busy for quite a while ;-).
Better get cracking then ... :-).
Oh, and it might be a good idea to put your machine in a Tooth Fairy proof enclosure whilst you're at it ;-).
Cheers,
Gary.
And when will the uploads of
)
And when will the uploads of past task completions resume?
The problem with the upload
)
The problem with the upload server was resolved some time ago and as far as I'm aware, all outstanding uploads were cleared soon afterwards. I don't know of any recurrence of a problem with uploading completed results.
When you say "past task completions" are you referring to tasks completed before the upload server was brought back on line - ie prior to 14th Jan?
Can you tell us what particular science run these tasks are from, when they were completed, and what messages you get in the BOINC event log when you try to upload them?
This seemed so strange that I just had a look through your host's task list and you only have 8 'in-progress' tasks, the oldest of which was sent to you around 24 hours ago. It's a bit hard to imagine you have any outstanding uploads. Perhaps you're referring to something else other than uploads so maybe you could rephrase your question?
Cheers,
Gary.
Anynoe know if there are
)
Anynoe know if there are other Boinc projects thar utilise the screen saver
function ? Seti And Einstein are already running. I would like to add a few more.[/i]