Given is a sequence of $n$ positive integers $w_1,w_2,\ldots,w_n$
that are associated with the items $1,2,\ldots,n$ respectively. In
the {\em windows scheduling} problem, the goal is to schedule all
the items (equal length information pages) on broadcasting
channels such that the gap between two consecutive appearances of
page $i$ on any of the channels is at most $w_i$ slots (a slot is
the transmission time of one page). In the {\em inverse integer bin
packing} problem, the goal is to pack all the items in bins of
unit size where item $i$ size (width) is $1/w_i$. The optimization
objective is to minimize the number of channels or bins. In the
off-line setting the sequence is known in advance whereas in the
on-line setting the items arrive in order and assignment decisions
are irrevocable. Since a page requires at least $1/w_i$ of the
channel's bandwidth, it follows that windows scheduling without
migration (all broadcasts of a page must be from the same channel)
is a restricted version of inverse integer bin packing.

Let $H=\ceil{\sum_{i=1}^{n}(1/w_i)}$ be the obvious bandwidth
lower bound on the required number of bins (channels). Previously
an $H+O(\ln H)$ off-line algorithm for the windows scheduling
problem was known. This paper presents an $H+1$ off-line algorithm
to the inverse integer bin packing problem. In the on-line
setting, this paper presents an $H+O(\sqrt{H})$ algorithm to both
problems where the one for the inverse integer bin packing problem
is simpler. On the other hand, this paper shows that already for
the inverse integer bin packing problem, any on-line algorithm
must use at least $H+\Omega(\ln H)$ bins.