A cyclic scheduling problem with an undetermined number of parallel identical processors

This paper presents two integer linear programming (ILP) models for cyclic scheduling of tasks with unit/general processing time. Our work is motivated by digital signal processing (DSP) applications on FPGAs (Field-Programmable Gate Arrays)—hardware architectures hosting several sets of identical arithmetic units. These hardware units can be formalized as dedicated sets of parallel identical processors. We propose a method to find an optimal periodic schedule of DSP algorithms on such architectures where the number of available arithmetic units must be determined by the scheduling algorithm with respect to the capacity of the FPGA circuit. The emphasis is put on the efficiency of the ILP models. We show the advantages of our models in comparison with common ILP models on benchmarks and randomly generated instances.

[1]  D. West.  Introduction to Graph Theory , 1995 .

[2]  Claire Hanen,et al.  Study of a NP-hard cyclic scheduling problem: The recurrent job-shop , 1994 .

[3]  B. Ramakrishna Rau,et al.  Some scheduling techniques and an easily schedulable horizontal architecture for high performance scientific computing , 1981, MICRO 14.

[4]  Sabih H. Gerez,et al.  A Genetic Approach to the Overlapped Scheduling of Iterative Data-Flow Graphs for Target Architectures with Communication Delays , 1997 .

[5]  Scott A. Mahlke,et al.  Increasing hardware efficiency with multifunction loop accelerators , 2006, Proceedings of the 4th International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS '06).

[6]  Peter Brucker,et al.  Tabu Search Algorithms for Cyclic Machine Scheduling Problems , 2005, J. Sched..

[7]  Miodrag Potkonjak,et al.  Fast prototyping of datapath-intensive architectures , 1991, IEEE Design & Test of Computers.

[8]  Premysl Sucha,et al.  Deadline constrained cyclic scheduling on pipelined dedicated processors considering multiprocessor tasks and changeover times , 2008, Math. Comput. Model..

[9]  Sabih H. Gerez,et al.  Range-chart-guided iterative data-flow graph scheduling , 1992 .

[10]  Sabih H. Gerez,et al.  An Integer Linear Programming Approach to the Overlapped Scheduling of Iterative Data-Flow Graphs for Target Architectures with Communication Delays , 2000 .

[11]  Dirk Fimmel,et al.  Optimal Software Pipelining Under Resource Constraints , 2001, Int. J. Found. Comput. Sci..

[12]  E. F. Girczyc,et al.  HAL: A Multi-Paradigm Approach to Automatic Data Path Synthesis , 1986, 23rd ACM/IEEE Design Automation Conference.

[13]  Alix Munier.  The complexity of a cyclic scheduling problem with identical machines and precedence constraints , 1996 .

[14]  A. Fettweis.  Wave digital filters: Theory and practice , 1986, Proceedings of the IEEE.

[15]  Wonyong Sung,et al.  Combined word-length optimization and high-level synthesis ofdigital signal processing systems , 2001, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[16]  Lars Wanhammar,et al.  Implementation of fast DSP algorithms using bit-serial arithmetic , 1994 .

[17]  Jirí Kadlec,et al.  Logarithmic Number System and Floating-Point Arithmetics on FPGA , 2002, FPL.

[18]  Dupont de Dinechin.  Time-Indexed Formulations and a Large Neighborhood Search for the Resource-Constrained Modulo Scheduling Problem , 2007 .

[19]  Guang R. Gao,et al.  A Polynomial Time Method for Optimal Software Pipelining , 1992, CONPAR.

[20]  Premysl Sucha,et al.  Scheduling of Iterative Algorithms with Matrix Operations for Efficient FPGA Design—Implementation of Finite Interval Constant Modulus Algorithm , 2007, J. VLSI Signal Process..

[21]  Claire Hanen,et al.  A Study of the Cyclic Scheduling Problem on Parallel Processors , 1995, Discret. Appl. Math..

[22]  Keshab K. Parhi,et al.  ILP-based cost-optimal DSP synthesis with module selection and data format conversion , 1998, IEEE Trans. Very Large Scale Integr. Syst..

[23]  Klaus Neumann,et al.  Truncated branch-and-bound, schedule-construction, and schedule-improvement procedures for resource-constrained project scheduling , 2001, OR Spectr..