One of my crunchers was due for some fresh work. I am used to getting periodic single downloads between 11 and 15mb. On April 7th, I received 3 block units. H1_0676.9, 1081.4, and 0762.9. 11.58, 14.05, and 11.58mb each. This is a first. I appreciate the work, but 3x? Has anyone else had this occur? Does Einstein think my cruncher is now three computers? My guess is that something is wrong with the list of active computers. I had a previous incident on another machine which received two work blocks. One was deleted the next day.
Advise as appropriate.
Copyright © 2024 Einstein@Home. All rights reserved.
Einstein Admin, Modem Users Beware-37meg Download
)
The current set of WUs is almost done, the Results currently given out to a large part consist of leftovers, so it may happen that there are data files around for which there is only one Result left to be crunched. Very bad luck that you apparently got three of them, but I'm afraid this will happen more frequently during the next week or so...
BM
BM
I wonder if there is any way
)
I wonder if there is any way the scheduler could detect who uses a modem connection and route the odds and ends to other participants -- ADSL users and those on a company / school LAN are usually better able to cope with a lot of large downloads than dial-ups; especially if the call is metered!
Maybe a new prefern#ences setting "Prefer to avoid large downloads" (Y/N)"
~~gravywavy
1. At the end of processing
)
1. At the end of processing such a predefined set of data files there comes a time where the whole pool of remaining Results consist of such WUs. This is unavoidable.
2. Coding and testig takes its time, we have not enough left for modifications to any of the components, and we are quite busy preparing the next search step.
3. What is a "large" download is quite relative, the size of the data files may also vary between the searches we perform. A simple switch, I guess, won't do.
=> for the time being I think we are better off keeping the user informed and let him decide what to do than sticking something together that takes the decision in a way he might not want to.
BM
BM
> => for the time being I
)
> => for the time being I think we are better off keeping the user informed
> and let him decide what to do than sticking something together that takes the
> decision in a way he might not want to.
>
> BM
>
fair comment. Thanks!
~~gravywavy