Prefix computations on a generalized mesh-connected computer with multiple buses

The mesh-connected computer with multiple buses (MC-CMB) is a well-known parallel organization, providing broadcast facilities in each row and each column. In this paper, we propose a 2D generalized MCCMB (2-GMCCMB) for the purpose of increasing the efficiency of executing some important applications of prefix computations such as solving Linear recurrences and tridiagonaI systems, etc. A k₁n₁ ×k₁n₂ 2-GMCCMB is constructed from a k ₁n₁×k₁n₂ mesh organization by enhancing the power of each disjoint n₁×n₂ submesh with multiple buses (sub-2-MCCMB). Given n data, a prefix computation can be performed in O(n^1/10) time on an n^3/5×n^2/5 2-GMCCMB, where each disjoint sub-2-MCCMB is of size n^1/2×n^3/10. This time bound is faster than the previous time bound of O(n^1/8) for the same computation on an n^5/8×n^3/8 2-MCCMB. Furthermore, the time bound of our parallel prefix algorithm can be further reduced to O(n^1/11) if fewer processors are used. Our result can be extended to the d-dimensional GMCCMB, giving a time bound of O(n¹ (d2(d)+d)/) for any constant d; here, we omit the constant factors. This time bound is less than the previous time bound of O(n¹(d2(d))/) on the d-dimensional MCCMB