I just signed up for Einstein@home, and was pleased to see a whole bunch of stuff for my SETI and LHC starved CPU to crunch. I think I have returned about 8 WUs by now, have received credit for one, but the others aren't even showing as pending after two days. There are no communication errors.... everything is running smoothly. The units being crunched are all SS5Ra_0
How long does it normally take with Einstein@home to get these on the board as pending?
Copyright © 2024 Einstein@Home. All rights reserved.
HELP - where are my completed jobs
)
Hi!
After a result is *reported*, it should appear within seconds on the web interface. However, for reasons of server resource conservation, so to speak, BOINC deliberately delays reporting finished jobs in order to report many finished results in a single batch. You can force reporting finished results by pressing the "Update" button in the advanced view of the BOINC manager GUI.
CU
Bikeman
Thank you Bikeman, but the
)
Thank you Bikeman, but the reporting is not the issue. They are going through the actual transfer, although I notice that they still show at 100% in my task list, as they do with other projects. I guess I'll wait and see if they show up in the "Your Results" area soon.
Now I just need to figure out how to get my signature to appear and for Boinc Stats to see that I'm actually in this project now (which was automatic whn I added LHC. I think I'll need to wait for that too.
Thanks again
Hey there....it worked! I
)
Hey there....it worked!
I spoke too soon...on to the signature now
Signature and Avatar test
)
Signature and Avatar test
RE: Thank you Bikeman, but
)
But you are seeing what he is saying....when a unit is completed by us crunchers a quick done note is sent to the Project but the actual unit results are not sent until either new work is needed for that project or we push the update button. This means it will show 100% on your list until one of those two things happens.
RE: when a unit is
)
Actually th result file gets *uploaded* physically to the project server reather quickly most of the time. But that is different from what is called "reporting" in BOINC-speak. An extra message is sent later to the project informing the project that the one or (most of the time) several results where uploaded since the last report and are ready to be "assimilated".
CU
Bikeman
RE: RE: when a unit is
)
So why the double reporting then? Can't the Projects computers figure out on their own that a result was returned and it then needs to have something done to it? If not it is just a computer, it just needs a programmer turned loose to do it. Double reporting seems to be wasting the computers time when it could be doing other things. Kind of like me telling you that your shoe is untied, and then repeating myself. Your brain knows you already know that, there is no need for me to tell you again.
Inquiring minds just want to know.
RE: RE: RE: when a unit
)
You seem to be using 2 terms (report and result upload) that have distinctly different meaning as if they mean the same thing. You MUST distinguish between the result upload and the report. When you have that difference clear in your mind then read his post again and you will see, hopefully, that it is NOT double reporting.
The result upload and the report happen separately for at least 2 reasons:
1) At some projects, results and reports go to different servers therefore the result and the report are necessarily sent separately.
2) Uploading the result is pretty simple stuff. Since uploadin a file requires no database access or other action, uploading 1 result at a time is no less efficient than uploading 2 or 3 results at a time. The reports are a different matter. They require database transactions and those are considerably more efficient if done 3 or more at a time rather than 1 at a time. Therefore the client trys to wait until you have several reports to send to a project, to reduce the load on the server.
Now, why can't the server figure out a result has been returned and that it needs something done to it? The upload is just an upload. The server accepts the upload without knowing what it is or who it came from, only that whoever uploaded it had authorization to do so. Period. That's all. The report that follows tells the server what the uploaded file is, who it came from, how much time was spent crunching it, how many credits are requested, etc. From that info, the server figures out what to do with the result.
BOINC FAQ Service
Official BOINC wiki
Installing BOINC on Linux