Parallel Sorting with Serial Memories

This correspondence examines the problem of sorting on a network of processors, where each processor consists of a single storage register and a small control unit capable of comparing two numbers and has a single serial memory attached to it. We show how to sort optimally on one- or two-dimensional arrays of p processors in time ¿(n + (n2/p2)) and ¿((n/¿p) + (n2/p2)), respectively. Because of the implementational advantages of serial memories, we feel that our architecture will be attractive for several applications.

[1]  Gérard M. Baudet,et al.  Optimal Sorting Algorithms for Parallel Computers , 1978, IEEE Transactions on Computers.

[2]  Fabrizio Luccio,et al.  Permutation of data blocks in a bubble memory , 1979, CACM.

[3]  Robert Michael Owens,et al.  New VLSI Architectures with Reduced Hardware , 1983 .

[4]  Chak-Kuen Wong,et al.  On the Complexity of Sorting in Magnetic Bubble Memory Systems , 1980, IEEE Transactions on Computers.

[5]  H. T. Kung,et al.  Sorting on a mesh-connected parallel computer , 1977, CACM.

[6]  David E. Muller,et al.  Bounds to Complexities of Networks for Sorting and for Switching , 1975, JACM.

[7]  Stephen Todd,et al.  Algorithm and Hardware for a Merge Sort Using Multiple Processors , 1978, IBM J. Res. Dev..

[8]  Thompson.  The VLSI Complexity of Sorting , 1983, IEEE Transactions on Computers.

[9]  Kenneth E. Batcher,et al.  Sorting networks and their applications , 1968, AFIPS Spring Joint Computing Conference.

[10]  Sartaj Sahni,et al.  Bitonic Sort on a Mesh-Connected Parallel Computer , 1979, IEEE Transactions on Computers.

[11]  Harold S. Stone,et al.  Parallel Processing with the Perfect Shuffle , 1971, IEEE Transactions on Computers.