NTHRYS › PhD Assistance › Parallel Computing

Parallel Computing

Field

▾

Parallel Computing

Select a category to explore research frontiers

Loading categories...

Research Frontiers in Automatic Loop Parallelization

Development of compiler techniques to automatically detect parallelizable loops and generate optimized parallel code with minimal programmer intervention.

Compiler-Driven Dependency Analysis in Irregular Loop Structures

Latency Hiding Through Speculative Loop Parallelization

Heterogeneous Accelerator Mapping for Nested Loop Optimization

Dynamic Loop Granularity Tuning at Runtime

Memory Coalescing in Auto-Parallelized Stencil Computations

Vectorization and Parallelization Synergy in Modern Compilers

Polyhedral Optimization for Control-Flow Dependent Loops

Adaptive Parallelism Selection Across CPU-GPU Boundaries

Data Locality Prediction in Automatically Parallelized Code

Parallelization Heuristics for Machine Learning Compiler Decisions

All Parallel Computing PhD categories