Interval partition with bounded overlap

Abstract The paper considers an optimization technique with applications to some resource-allocation problems that arise in high-level synthesis of digital systems. In an abstract sense, the optimization problem is that of the partitioning of a set of line intervals into a minimum number of subsets such that intervals assigned to a subset satisfy certain properties. It is shown that the problem can be solved optically in O(n) time, where n is the number of line intervals. Applications of the problem to memory allocation and task scheduling are discussed. Specifically, a program called MEMAL is described for the allocation of variables to multiport memories. MEMAL has been implemented in PASCAL on a Sun/3 workstation.

[1]  Alfred V. Aho,et al.  The Design and Analysis of Computer Algorithms , 1974 .

[2]  M. Golumbic.  Algorithmic graph theory and perfect graphs , 1980 .

[3]  C. V. Ramamoorthy,et al.  Optimal Scheduling Strategies in a Multiprocessor System , 1972, IEEE Transactions on Computers.

[4]  Arun K. Majumdar,et al.  Allocation of multiport memories in data path synthesis , 1988, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[5]  Joseph Y.-T. Leung,et al.  An Optimal Solution for the Channel-Assignment Problem , 1979, IEEE Transactions on Computers.

[6]  Daniel P. Siewiorek,et al.  Automated Synthesis of Data Paths in Digital Systems , 1986, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[7]  Les G. Proll,et al.  Algorithms: Algorithm 341: solution of linear programs in 0-1 variables by implicit enumeration , 1968, CACM.