Compiling Fortran D for MIMD distributed-memory machines
Communications of the ACM
Dynamic control of performance monitoring on large scale parallel systems
ICS '93 Proceedings of the 7th international conference on Supercomputing
OpenMP: An Industry-Standard API for Shared-Memory Programming
IEEE Computational Science & Engineering
Loop-Level Parallelism in Numeric and Symbolic Programs
IEEE Transactions on Parallel and Distributed Systems
TEST: a tracer for extracting speculative threads
Proceedings of the international symposium on Code generation and optimization: feedback-directed and runtime optimization
LLVM: A Compilation Framework for Lifelong Program Analysis & Transformation
Proceedings of the international symposium on Code generation and optimization: feedback-directed and runtime optimization
Pin: building customized program analysis tools with dynamic instrumentation
Proceedings of the 2005 ACM SIGPLAN conference on Programming language design and implementation
POSH: a TLS compiler that exploits program structure
Proceedings of the eleventh ACM SIGPLAN symposium on Principles and practice of parallel programming
SPEC CPU2006 benchmark descriptions
ACM SIGARCH Computer Architecture News
Valgrind: a framework for heavyweight dynamic binary instrumentation
Proceedings of the 2007 ACM SIGPLAN conference on Programming language design and implementation
Software behavior oriented parallelization
Proceedings of the 2007 ACM SIGPLAN conference on Programming language design and implementation
Compiler-Driven Dependence Profiling to Guide Program Parallelization
Languages and Compilers for Parallel Computing
Alchemist: A Transparent Dependence Distance Profiling Infrastructure
Proceedings of the 7th annual IEEE/ACM International Symposium on Code Generation and Optimization
SD3: A Scalable Approach to Dynamic Data-Dependence Profiling
MICRO '43 Proceedings of the 2010 43rd Annual IEEE/ACM International Symposium on Microarchitecture
| Hi-index | 0.00 |
Compiler-based automatic parallelization has been studied for many years. Static dependence analysis has played an important role, as dynamic analysis requires massive computing power. As multicore processors are widely used and computation power of processors has increased dramatically, dynamic data dependence analysis is gaining popularity. However, its time and memory overhead is still a big burden. Researchers investigated techniques to minimize these overheads, while they find data dependences in target programs. In this paper, we present CUBIT, a dynamic dependence analysis tool that identifies potential parallel loop candidates. CUBIT requires only small memory footprint by using bitmap profiling and takes advantage of SIMD optimized algorithm. As a result, CUBIT successfully minimizes analysis time and memory overhead for dynamic data dependence analysis.