A set of input vectors S is conclusive if correct functionality for all input vectors is implied by correct functionality over vectors in S. We consider four functionalities of comparator networks: sorting, merging of two equal length sorted vectors, sorting of bitonic vectors, and halving (i.e., separating values above and below the median). For each of… (More)
This paper points to an abnormal phenomena of comparator networks. For most key processing problems (such as sorting, merging or insertion) the smaller the input size the easier the problem. Surprisingly, this is not the case for Bitonic sorting. Namely, the minimal depth of a comparator network that sorts all Bitonic sequences of n keys is not monotonic in… (More)
This work studies comparator networks in which several of the outputs are accelerated. That is, they are generated much faster than the other outputs, and this without hindering the other outputs. We study this acceleration in the context of merging networks and sorting networks. The paper presents a new merging technique, the Tri-section technique, that… (More)
The ¼-½ Principle of Knuth and its many variations are well-known in the context of comparator networks. However, comparator networks is not the strongest model of computation obeying this principle. We present another model of computation that obeys all known ¼-½ principles. Moreover, it is the strongest model obeying some variants of the ¼-½ Principle.
Building on previous works, this paper establishes that the minimal depth of a Bitonic sorter of n keys is 2 log(n) − log(n).
This paper studies fast Bitonic sorters of arbitrary width. It constructs such a sorter of width n and depth log(n) + 3, for any n (not necessarily a power of two).