CGO 2012
26 papers
- An automatic code overlaying technique for multicores with explicitly-managed memory hierarchies
- Auto-generation and auto-tuning of 3D stencil codes on GPU clusters
- Automatic speculative DOALL for clusters
- Compiling for automatically generated instruction set extensions
- Compiling for niceness: mitigating contention for QoS in warehouse scale computers
- DeadSpy: a tool to pinpoint program inefficiencies
- Deferred methods: accelerating dynamic program analysis on multicores
- Dynamic compilation of data-parallel kernels for vector processors
- Dynamically managed data for CPU-GPU architectures
- Efficient and accurate data dependence profiling using software signatures
- Efficient bottom-up heap analysis for symbolic path-based data access summaries
- HELIX: automatic parallelization of irregular programs for chip multiprocessing
- HQEMU: a multi-threaded and retargetable dynamic binary translator on multicores
- Hierarchical overlapped tiling
- Light-weight bounds checking
- Matching memory access patterns and data placement for NUMA systems
- Micro-specialization: dynamic code specialization of database management systems
- On-demand dynamic summary-based points-to analysis
- Panacea: towards holistic optimization of MapReduce applications
- Phase guided profiling for fast cache modeling
- PinADX: an interface for customizable debugging with dynamic instrumentation
- Reconciling transactional conflicts with compiler's help
- Runtime asynchronous fault tolerance via speculation
- Scan detection and parallelization in "inherently sequential" nested loop programs
- Using graph-based program characterization for predictive modeling
- WCET-aware static locking of instruction caches