load based handover

[Neels Hofmeyr]

Hi Jolly,

my current task is to implement load based handover between BTS, and in old
tradition of adopting your code to implement dynamic timeslots on
osmo-{bts,bsc}, I am now looking at the openbsc.git jolly/new_handover branch.

I notice that, on that branch, there are improvements on the "old" handover,
followed by a new elaborate handover_2 algorithm, which looks very sane at
first sight, paired with an elaborate test suite.

So my question is whether handover_2.c has some obvious shortcomings or open
issues that need to be resolved before I can consider merging that to OsmoBSC's
master branch.

Should we still keep the "old" handover decision code around as well for
particular reasons, or would it make sense to adopt handover_2.c by default?
Is one better than the other in particular situations?

And ... do you remember any high level conclusions from when you implemented
the branch, things I should be aware of when testing and adopting the code? Did
it work out as expected? How well tested is it?

Thanks!

~N
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://lists.osmocom.org/pipermail/openbsc/attachments/20171124/65340694/attachment.bin>