Automatic data movement and computation mapping for multi-level parallel architectures with explicitly managed memories
Muthu Manikandan Baskaran, Uday Kumar Reddy Bondhugula, Sriram Krishnamoorthy et al. · ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming · 2008
Several parallel architectures such as GPUs and the Cell processor have fast explicitly managed on-chip memories, in addition to slow off-chip memory. They also have very high computational power with multiple levels of parallelism. A significant challenge in programming these architectures is to effectively exploit the parallelism available in the architecture and manage the fast memories to maximize performance.
Cited by 112 works. doi:10.1145/1345206.1345210
Authors
- Muthu Manikandan Baskaran
- Uday Kumar Reddy Bondhugula
- Sriram Krishnamoorthy
- Jagannathan Ramanujam
- Atanas Rountev
- Ponnuswamy Sadayappan
References
- Nonlinear Programming.
- Automatically Tuned Linear Algebra Software
- Optimizing matrix multiply using PHiPAC
- The Omega test: a fast and practical integer programming algorithm for dependence analysis
- A practical algorithm for exact array dependence analysis
- Optimizing matrix multiply using PHiPAC
- Dataflow analysis of array and scalar references
- Memory---Sequoia
- Some efficient solutions to the affine scheduling problem. I. One-dimensional time
- Parametric integer programming