SHIFT is a topology-agnostic dynamic compute relocation framework for chiplet-based systems that reports 75.2-97.9% successful relocations, up to 76.8% latency reduction, 12.5x throughput gain, and 4.9x/5.9x/1.8x improvements in runtime/throughput/energy on LLM workloads.
Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SHIFT: Dynamic Compute Relocation Framework for Communication-Aware Chiplet-Based Systems
SHIFT is a topology-agnostic dynamic compute relocation framework for chiplet-based systems that reports 75.2-97.9% successful relocations, up to 76.8% latency reduction, 12.5x throughput gain, and 4.9x/5.9x/1.8x improvements in runtime/throughput/energy on LLM workloads.