Finding Coarse Grained Parallelism in Computational Geometry Algorithms
نویسنده
چکیده
A technique, permitting automatic finding coarse grained parallelism in algorithms presented with arbitrary nested loops, is presented. The technique is based on finding affine space partition mappings. The main advantage of this technique is that it allows us to form constraints for finding mappings directly in a linear form while known techniques result in building non-linear constraints which should next be linearized. After finding affine space partition mappings, well-known code generation approaches can be applied to expose algorithm parallelism. It is shown how this technique can be applied for parallelizing computational geometry algorithms by means of two examples.
منابع مشابه
Feasibility of detecting and localizing radioactive source using image processing and computational geometry algorithms
We consider the problem of finding the localization of radioactive source by using data from a digital camera. In other words, the camera could help us to detect the direction of radioactive rays radiation. Therefore, the outcome could be used to command a robot to move toward the true direction to achieve the source. The process of camera data is performed by using image processing and computa...
متن کاملExposing Fine - Grained Parallelism in Algebraic Multigrid
Algebraic multigrid methods for large, sparse linear systems are a necessity in many computational simulations, yet parallel algorithms for such solvers are generally decomposed into coarse-grained tasks suitable for distributed computers with traditional processing cores. However, accelerating multigrid methods on massively parallel throughput-oriented processors, such as graphics processing u...
متن کاملAutomatic Discovery of Coarse-Grained Parallelism in Media Applications
With the increasing use of multi-core microprocessors and hardware accelerators in embedded media processing systems, there is an increasing need to discover coarse-grained parallelism in media applications written in C and C++. Common versions of these codes use a pointer-heavy, sequential programming model to implement algorithms with high levels of inherent parallelism. The lack of automated...
متن کاملExposing Fine-Grained Parallelism in Algebraic Multigrid Methods
Algebraic multigrid methods for large, sparse linear systems are a necessity in many computational simulations, yet parallel algorithms for such solvers are generally decomposed into coarse-grained tasks suitable for distributed computers with traditional processing cores. However, accelerating multigrid on massively parallel throughput-oriented processors, such as the GPU, demands algorithms w...
متن کاملExploiting Coarse-Grain Parallelism in the MPEG-2 Algorithm
As the demand for multimedia applications increases, the performance of algorithms such as MPEG-2 video compression on general-purpose microprocessors is becoming more important. In this paper, we propose a number of coarse-grained parallel implementations of MPEG-2 decoding and encoding. We evaluate the performance of these implementations on a single-chip multiprocessor, and compare the perfo...
متن کامل