Deep Jam: Conversion of Coarse-Grain Parallelism to Fine-Grain and Vector Parallelism
Explore this paper's citation graph
Summary
This paper proposes a feedback-directed deep jam algorithm, selecting a jamming strategy, function of the architecture and application characteristics, and contributes to the extraction of fine-grain parallelism in irregular applications.
- Type
- preprint
- Published
- 2007-01-01
- Cited by
- 2
- References
- 62
- OpenAlex
- https://openalex.org/W40503131
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:18661707
Keywords
Computer science, Parallel computing, Instruction-level parallelism, Control flow, Data parallelism
References
- Software pipelining: an effective scheduling technique for VLIW machines
- Hybrid Analysis: Static & Dynamic Memory Reference Analysis
- Optimizing Compilers for Modern Architectures
- Adaptive Optimizing Compilers for the 21st Century
- Maximal Static Expansion
- The potential of data value speculation to boost ILP
- Efficiently computing static single assignment form and the control dependence graph
- Array-data flow analysis and its use in array privatization
- Special Issue on Program Generation, Optimization, and Platform Adaptation
- Integrated predicated and speculative execution in the IMPACT EPIC architecture
- Complementing software pipelining with software thread integration
- Single-dimension software pipelining for multi-dimensional loops
- The superblock: An effective technique for VLIW and superscalar compilation
- The LRPD test: speculative run-time parallelization of loops with privatization and reduction parallelization
- Array SSA form and its use in parallelization
- Software thread integration for embedded system display applications
- Improving software pipelining with unroll-and-jam
- A fast bit-vector algorithm for approximate string matching based on dynamic programming
- High-level synthesis of nonprogrammable hardware accelerators
- CHIMAERA: a high-performance architecture with a tightly-coupled reconfigurable functional unit
Cited by
- Étude de transformations et d'optimisations de code parallèle statique ou dynamique pour architecture "many-core". (Study of transformations and static or dynamic parallel code optimization for manycore architecture)
- Team Alchemy Architectures, Languages and Compilers to Harness the End of Moore Years
Related papers
- Parallelism viewpoint: An architecture viewpoint to model parallelism behaviour of parallelism-intensive software systems
- Relating Data-Parallelism and (and-) Parallelism in Logic Programs
- Exploiting task and data parallelism on a multicomputer
- A Model of Speculative Parallelism
- On the Concepts of Parallelism in Biomolecular Computing
- An Analytical study of Performance towards Task-level Parallelism on Many-core systems using Java API
- How much parallelism is there in irregular applications