arXiv:2606. 01680v1 Announce Type: cross Abstract: Network failures are among the most frequent hardware faults in large-scale GPU clusters and a leading cause of training-job interruptions.
arXiv:2607. 13332v1 Announce Type: new Abstract: Training large language models at the multi-billion to trillion parameter scale is confined to datacenters, where data-parallel (DP) and model-parallel (MP) techniques presume homogeneous accelerators, high-speed interconnects, and a single orchestrating entity.
arXiv:2609.13585v1 Announce Type: cross Abstract: Communication has become a bottleneck in distributed training and inference of large models. Overlapping communication with computation at the granul...
State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack. Existing fault-tolerance mechanisms either impose non-trivial overhead during failure-free execution or suffer from prolonged recovery latency, particularly under scenarios where a small subset of compute nodes experience permanent failures.
arXiv:2607. 01646v2 Announce Type: replace Abstract: State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack.
arXiv:2607. 01646v1 Announce Type: new Abstract: State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack.
The paper introduces REACT, a system that dynamically tunes communication collectives in distributed AI training to mitigate congestion without requiring network infrastructure changes. REACT operates at the application layer, detecting congestion via flow statistics and adjusting the pattern of data exchange—such as selecting different aggregation nodes in an AllReduce tree—while preserving the semantics of the communication. Evaluations on a shared academic GPU cluster show that REACT improves algorithm bandwidth by 13%–38% under congestion, with simulations indicating potential gains up to 75%.
arXiv:2607. 28633v1 Announce Type: cross Abstract: Disaggregated LLM inference creates a datacenter networking problem that no existing system solves correctly.
arXiv:2609.40093v1 Announce Type: cross Abstract: Expert parallelism (EP) enables inference of large Mixture-of-Experts (MoE) models by placing their experts across multiple GPUs, but requires substa...
arXiv:2402. 09589v2 Announce Type: replace-cross Abstract: We present MLCC, a novel technique to augment today's congestion control algorithms to accelerate DNN training jobs in shared GPU clusters in a fully distributed manner.
arXiv:2607. 23264v1 Announce Type: cross Abstract: Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and overlap data movement with Tensor Core computation.
arXiv:2606. 30919v1 Announce Type: cross Abstract: Edge-cloud inference collaborations are often designed with a routing estimator that decides whether to offload each frame from weak models at the edge to stronger models in the cloud.
arXiv:2607. 02043v1 Announce Type: cross Abstract: Disaggregated LLM serving runs prefill and decode on separate GPU pools to keep the two phases from interfering.