DICE achieves 1.77-1.90x dynamic energy efficiency and 42-46% power reduction versus modeled NVIDIA Turing SMs by executing SIMT workloads on pipelined CGRAs with p-graphs handling dynamism and 68% fewer register file accesses.
Optimizing NUCA organizations and wiring alternatives for large caches with CACTI 6.0,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DICE: Enabling Efficient General-Purpose SIMT Execution with Statically Scheduled Coarse-Grained Reconfigurable Arrays
DICE achieves 1.77-1.90x dynamic energy efficiency and 42-46% power reduction versus modeled NVIDIA Turing SMs by executing SIMT workloads on pipelined CGRAs with p-graphs handling dynamism and 68% fewer register file accesses.