An Evaluation of Coarse Grain Data
نویسندگان
چکیده
In this paper we present top down and bottom up methods for generating coarse grain dataaow or mul-tithreaded code, and evaluate their eeectiveness. The top down technique generates clusters directly from the intermediate data dependence graph used for compiler optimizations. Bottom up techniques coalesce ne-grain dataaow code into clusters. We measure the resulting number of clusters executed, cluster size, and number of inputs per cluster, for Livermore and Pur-due benchmarks. The top down method executes less clusters and instructions, but incurs a higher number of matches per cluster, which exempliies the need for eecient matching of more than two inputs per cluster.
منابع مشابه
Uncertainties in Evaluation of the Sediment Transport Rates in Typical Coarse-Bed Rivers in Iran
Flow and sediment transport processes are different and more complex in coarse-bed rivers than in sand-bed rivers. The main goal of the present study is to evaluate different modes of sediment transport using different hydrometric and hydraulic methods, and to address the major uncertainties. Four river reaches were selected as representatives of coarse-bed rivers in the Northwest of Iran. A se...
متن کاملCorrosion Inhibition of Sodium Phosphate for Coarse and Near Ultrafined-Grain Mild steel surface
An ultrafine grain surface layer with average crystallite size of 28 nm was produced on annealed mild steel through a wire brushing process. The effects of grain size reduction on the inhibition performance of sodium phosphate were investigated using polarization and electrochemical impedance spectroscopy (EIS) measurements. The crystal grain size of wire brushed surface was analyzed by X-ray d...
متن کاملParallel Object Instruction Scheduler
We describe an approach to parallel compilation that seeks to harness the vast amount of ne-grain parallelism that is exposed through partial evaluation of numerically-intensive scientiic programs. We have constructed a compiler for the Supercomputer Toolkit parallel processor that uses partial evaluation to break down data abstractions and program structure, producing huge basic blocks that co...
متن کاملPh.d. Dissertation Evaluation and Exploitation of Locality in the Data Driven Execution Model
OF PH.D. DISSERTATION EVALUATION AND EXPLOITATION OF LOCALITY IN THE DATA DRIVEN EXECUTION MODEL The advent of hybrid von Neumann-data driven architectures arose from a desire to combine the most salient features of coarse grain von Neumann and ne-grain data driven models. Hybrid architectures achieve high performance through concurrent execution and the exploitation of program and data localit...
متن کاملDataflow based Near-Data Processing using Coarse Grain Reconfigurable Logic
The emergence of 3D-DRAM has rekindled interest in near-data processing research. This article introduces and describes a method for near-data processing using hybrid dataflow techniques implemented with coarse grain reconfigurable logic. We provide an initial evaluation of the concept to justify further development. The melding of these technologies produces increased throughput while dramatic...
متن کامل