On Sat, 2007-03-10 at 07:55 +0100, David Nielsen wrote: > Do we know how many packages on average that would apply to, it seems to > me that unless this actually saves us bandwidth in general cases rather > than just OpenOffice.org (which is big and hideous by nature) then it > might not be worth it for anything but cool value. > > Has this been tried on something like the FC6 updates? Additionally what > are the impacts client side of doing this, I assume a certain overhead > in stitching the packages together is present. > > Aside that the idea sounds really neat and once it's ready for testing > I'd love to give it a go. > > - David Nielsen > I have a local repository with all of Core + some Extras + some Livna + some FreshRPMS in it. If someone had installed everything from my repository when I first created it, and then had do an update to the latest version of everything, they would have to download ~1.5 GB of updates. If that same person had the yum-deltarpm plugin enabled, they would only have to download 638 MB. That is *including* all the updates that don't have drpms because the savings isn't enough. That means that we're looking at a savings of roughly 60%. As for the rpm rebuild time, it's not that bad. Most of the time seems to be reading the data from the hard drive. Hope that answers your questions. Jonathan
Attachment:
signature.asc
Description: This is a digitally signed message part
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
