DCC co-optimizes data rearrangements and compute schedules for ML kernels on PIM memory, reporting up to 13.17x simulated speedup over a GPU-only baseline.
Darwin: A DRAM-Based Multi-Level Processing-in-Memory Architecture for Column-Oriented Database
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DCC: Data-Centric Compilation of Machine Learning Kernels for Processing-In-Memory Architectures
DCC co-optimizes data rearrangements and compute schedules for ML kernels on PIM memory, reporting up to 13.17x simulated speedup over a GPU-only baseline.