Tile size selection using cache organization and data layout
PLDI '95 Proceedings of the ACM SIGPLAN 1995 conference on Programming language design and implementation
Optimizing compilers for modern architectures: a dependence-based approach
Optimizing compilers for modern architectures: a dependence-based approach
The Importance of Prepass Code Scheduling for Superscalar and Superpipelined Processors
IEEE Transactions on Computers
CC '92 Proceedings of the 4th International Conference on Compiler Construction
A comparison of empirical and model-driven optimization
PLDI '03 Proceedings of the ACM SIGPLAN 2003 conference on Programming language design and implementation
FOCS '99 Proceedings of the 40th Annual Symposium on Foundations of Computer Science
Measuring Cache and TLB Performance and Their Effect of Benchmark Run
Measuring Cache and TLB Performance and Their Effect of Benchmark Run
Automatic Blocking of Nested Loops
Automatic Blocking of Nested Loops
Automatic measurement of memory hierarchy parameters
SIGMETRICS '05 Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
JuliusC: a practical approach for the analysis of divide-and-conquer algorithms
LCPC'04 Proceedings of the 17th international conference on Languages and Compilers for High Performance Computing
Combining analytical and empirical approaches in tuning matrix transposition
Proceedings of the 15th international conference on Parallel architectures and compilation techniques
Automatic performance model construction for the fast software exploration of new hardware designs
CASES '06 Proceedings of the 2006 international conference on Compilers, architecture and synthesis for embedded systems
A comparison of online and offline strategies for program adaptation
ACM-SE 45 Proceedings of the 45th annual southeast regional conference
Rapidly Selecting Good Compiler Optimizations using Performance Counters
Proceedings of the International Symposium on Code Generation and Optimization
Positivity, posynomials and tile size selection
Proceedings of the 2008 ACM/IEEE conference on Supercomputing
Automating the generation of composed linear algebra kernels
Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis
Automatic creation of tile size selection models
Proceedings of the 8th annual IEEE/ACM international symposium on Code generation and optimization
Parallel memory prediction for fused linear algebra kernels
ACM SIGMETRICS Performance Evaluation Review - Special issue on the 1st international workshop on performance modeling, benchmarking and simulation of high performance computing systems (PMBS 10)
Probabilistic auto-tuning for architectures with complex constraints
Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era
Analytical bounds for optimal tile size selection
CC'12 Proceedings of the 21st international conference on Compiler Construction
A script-based autotuning compiler system to generate high-performance CUDA code
ACM Transactions on Architecture and Code Optimization (TACO) - Special Issue on High-Performance Embedded Architectures and Compilers
| Hi-index | 0.00 |
A key step in program optimization is the determination of optimal values for code optimization parameters such as cache tile sizes and loop unrolling factors. One approach, which is implemented in most compilers, is to use analytical models to determine these values. The other approach, used in library generators like ATLAS, is to perform a global empirical search over the space of parameter values.Neither approach is completely suitable for use in general-purpose compilers that must generate high quality code for large programs running on complex architectures. Model-driven optimization may incur a performance penalty of 10-20% even for a relatively simple code like matrix multiplication. On the other hand, global search is not tractable for optimizing large programs for complex architectures because the optimization space is too large.In this paper, we advocate a methodology for generating high-performance code without increasing search time dramatically. Our methodology has three components: (i) modeling, (ii) local search, and (iii) model refinement. We demonstrate this methodology by using it to eliminate the performance gap between code produced by a model-driven version of ATLAS described by us in prior work, and code produced by the original ATLAS system using global search.