Mapping Linear Recurrences onto Systolic Arrays
نویسندگان
چکیده
We present an automatic method for mapping a system of linear recurrence equations onto systolic architectures. First, we show that systolic architectures can be derived from linear recurrence equations using the notion of directed recurrence equations. Next, we provide a procedure called CUBIZATION to achieve better performance while mapping such equations. The CUBIZATION procedure is completely automated and can be implemented to design systolic programs for a very general architecture referred to as Basic Systolic Architecture (BSA). Using BSA, we obtain specific target architectures. The method is illustrated using a running example of Gauss-Jordan diagonalization.
منابع مشابه
University of Groningen Mapping systolic FIR filter banks onto fixed-size linear processor arrays
A technique for mapping systolic FIR filter banks onto fixed-size processor arrays is presented. It is based on the time-sharing properties of c-slow circuits. The technique can be further developed to a formalism and holds high potential for automatic realization. It has been applied to the mapping of systolic filter banks onto a fixed-size array of Transputers.
متن کاملMapping of Neural Network Models onto Systolic Arrays
This paper presents a mapping scheme for the proposed implementation of neural network models on systolic arrays. The mapping technique is illustrated on the multilayer perceptron with back-propagation learning. Dependency graphs have been given that represent the operations in the execution phases of the neural network model and later suitable algorithms are presented to realize the operations...
متن کاملSystematic Methodology of Mapping Signal Processing Algorithms into Arrays of Processors
Nowadays high speed signal processing has become the only alternative in modern communication system, given the rapidly growing microelectronics technology. This high speed, real time signal processing depends critically both on the parallel algorithms and on parallel processor technology. Special purpose array processor structures will have become the real possibility for high speed signal pro...
متن کاملAn E cient Allocation Strategy for MappingA ne Recurrences into
This paper adresses the problem of eecient mappings of nested loops, and more generally of systems of aane recurrence equations, into regular arrays. The presented technique is based on the transformation of an initial systolic mapping. By studying the processor element (PE) activity, a nearly space-optimal mapping is designed by serializing the computations of several initial PEs into a single...
متن کاملA New Formulation of the Mapping Conditionsfor the Synthesis of Linear Systolic
This paper presents a new formulation for mapping algorithms into linear systolic arrays. The closed-form necessary and suucient mapping conditions are derived to identify mappings without computation connicts and data link collisions. These mapping conditions are easy to check because their constituent variables are the space-time mapping matrix and the problem size parameters. The design of o...
متن کامل