CAIS delivers 1.38x end-to-end LLM training speedup over NVLS and 1.61x over T3 by making in-switch computing aware of computation memory requirements instead of treating communication as an isolated phase.
Accel-sim: An extensible simulation framework for validated gpu modeling,
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
MoE-Hub enables seamless MoE communication overlap via hardware-accelerated destination-agnostic data transmission, delivering 1.40x-3.08x per-layer and 1.21x-1.98x end-to-end speedups over prior systems.
DySHARP accelerates MoE expert parallelism via dynamic multimem addressing and token-centric kernel fusion to cut redundant traffic and deliver up to 1.79x speedup over prior in-switch solutions.
StableHLO serves as a viable unified representation for cross-architecture performance modeling of distributed ML workloads, preserving relative trends while exposing fidelity trade-offs.
citing papers explorer
-
Towards Compute-Aware In-Switch Computing for LLMs Tensor-Parallelism on Multi-GPU Systems
CAIS delivers 1.38x end-to-end LLM training speedup over NVLS and 1.61x over T3 by making in-switch computing aware of computation memory requirements instead of treating communication as an isolated phase.
-
MoE-Hub: Taming Software Complexity for Seamless MoE Overlap with Hardware-Accelerated Communication on Multi-GPU Systems
MoE-Hub enables seamless MoE communication overlap via hardware-accelerated destination-agnostic data transmission, delivering 1.40x-3.08x per-layer and 1.21x-1.98x end-to-end speedups over prior systems.
-
Accelerating MoE with Dynamic In-Switch Computing on Multi-GPUs
DySHARP accelerates MoE expert parallelism via dynamic multimem addressing and token-centric kernel fusion to cut redundant traffic and deliver up to 1.79x speedup over prior in-switch solutions.
-
Evaluating Cross-Architecture Performance Modeling of Distributed ML Workloads Using StableHLO
StableHLO serves as a viable unified representation for cross-architecture performance modeling of distributed ML workloads, preserving relative trends while exposing fidelity trade-offs.