arXiv subjects

Shengwei Li

Publications and source records attributed to Shengwei Li.

7 recordsLinked to original sources

arXiv2025

H2:Towards Efficient Large-Scale LLM Training on Hyper-Heterogeneous Cluster over 1,000 Chips

Recent advancements in large language models (LLMs) necessitate extensive computational resources, prompting the use of diverse hardware accelerators from multiple vendors. However, traditional distributed training frameworks struggle to efficiently utilize hyper-heterogeneous clusters comprising thousands of chips due to significant disparities in software stacks, operator implementations, communication libraries, and hardware capabilities. To address these challenges, we propose H2, which stands for HyperHetero and is a systematic framework enabling efficient training of LLMs on clusters with over 1,000 heterogeneous chips. H2 incorporates DiTorch, a unified PyTorch-compatible interface ensuring program consistency across chips, and DiComm, a device-direct RDMA communication library optimized for heterogeneous environments. Furthermore, we introduce HeteroPP with HeteroAuto, an adaptive pipeline parallelism strategy that dynamically balances computational load, memory limitations, and communication overhead. Evaluations on a 100-billion-parameter LLM demonstrate that our approach consistently achieves a superlinear speedup, outperforming baseline homogeneous training solutions by up to 16.37% in our experiments. These findings validate the feasibility and efficiency of hyper-heterogeneous training at unprecedented scales.

cs.DC↗

arXiv2024

Pro-Prophet: A Systematic Load Balancing Method for Efficient Parallel Training of Large-scale MoE Models

The size of deep learning models has been increasing to enhance model quality. The linear increase in training computation budget with model size means that training an extremely large-scale model is exceedingly time-consuming. Recently, the Mixture of Expert (MoE) has drawn significant attention as it can scale models to extra-large sizes with a stable computation budget. However, inefficient distributed training of large-scale MoE models hinders their broader application. Specifically, a considerable dynamic load imbalance occurs among devices during training, significantly reducing throughput. Several load-balancing works have been proposed to address the challenge. System-level solutions draw more attention for their hardware affinity and non-disruption of model convergence compared to algorithm-level ones. However, they are troubled by high communication costs and poor communication-computation overlapping. To address these challenges, we propose a systematic load-balancing method, Pro-Prophet, which consists of a planner and a scheduler for efficient parallel training of large-scale MoE models. To adapt to the dynamic load imbalance, we profile training statistics and use them to design Pro-Prophet. For lower communication volume, Pro-Prophet planner determines a series of lightweight load-balancing strategies and efficiently searches for a communication-efficient one for training based on the statistics. For sufficient overlapping of communication and computation, Pro-Prophet scheduler schedules the data-dependent operations based on the statistics and operation features, further improving the training throughput. Experimental results indicate that Pro-Prophet achieves up to 2.66x speedup compared to Deepspeed-MoE and FasterMoE. Additionally, Pro-Prophet achieves a load-balancing enhancement of up to 11.01x when compared to FasterMoE.

cs.DC↗

arXiv2023

Towards Understanding the Generalizability of Delayed Stochastic Gradient Descent

Stochastic gradient descent (SGD) performed in an asynchronous manner plays a crucial role in training large-scale machine learning models. However, the generalization performance of asynchronous delayed SGD, which is an essential metric for assessing machine learning algorithms, has rarely been explored. Existing generalization error bounds are rather pessimistic and cannot reveal the correlation between asynchronous delays and generalization. In this paper, we investigate sharper generalization error bound for SGD with asynchronous delay $\tau$. Leveraging the generating function analysis tool, we first establish the average stability of the delayed gradient algorithm. Based on this algorithmic stability, we provide upper bounds on the generalization error of $\tilde{\mathcal{O}}(\frac{T-\tau}{n\tau})$ and $\tilde{\mathcal{O}}(\frac{1}{n})$ for quadratic convex and strongly convex problems, respectively, where $T$ refers to the iteration number and $n$ is the amount of training data. Our theoretical results indicate that asynchronous delays reduce the generalization error of the delayed SGD algorithm. Analogous analysis can be generalized to the random delay setting, and the experimental results validate our theoretical findings.

cs.LG↗

arXiv2023

Oases: Efficient Large-Scale Model Training on Commodity Servers via Overlapped and Automated Tensor Model Parallelism

Deep learning is experiencing a rise in large-scale models. Training large-scale models is costly, prompting researchers to train large-scale models on commodity servers that more researchers can access. The massive number of parameters necessitates the use of model parallelism training methods. Existing studies focus on training with pipeline model parallelism. However, the tensor model parallelism (TMP) is inevitable when the model size keeps increasing, where frequent data-dependent communication and computation operations significantly reduce the training efficiency. In this paper, we present Oases, an automated TMP method with overlapped communication to accelerate large-scale model training on commodity servers. Oases proposes a fine-grained training operation schedule to maximize overlapping communication and computation that have data dependence. Additionally, we design the Oases planner that searches for the best model parameter partition strategy of TMP to achieve further accelerations. Unlike existing methods, Oases planner is tailored to model the cost of overlapped communication-computation operations. We evaluate Oases on various model settings and two commodity clusters, and compare Oases to four state-of-the-art implementations. Experimental results show that Oases achieves speedups of 1.01--1.48$\times$ over the fastest baseline, and speedups of up to 1.95$\times$ over Megatron.

cs.DC↗

arXiv2022

Merak: An Efficient Distributed DNN Training Framework with Automated 3D Parallelism for Giant Foundation Models

Foundation models are becoming the dominant deep learning technologies. Pretraining a foundation model is always time-consumed due to the large scale of both the model parameter and training dataset. Besides being computing-intensive, the training process is extremely memory-intensive and communication-intensive. These features make it necessary to apply 3D parallelism, which integrates data parallelism, pipeline model parallelism and tensor model parallelism, to achieve high training efficiency. To achieve this goal, some custom software frameworks such as Megatron-LM and DeepSpeed are developed. However, current 3D parallelism frameworks still meet two issues: i) they are not transparent to model developers, which need to manually modify the model to parallelize training. ii) their utilization of computation, GPU memory and network bandwidth are not sufficient. We propose Merak, an automated 3D parallelism deep learning training framework with high resource utilization. Merak automatically deploys with an automatic model partitioner, which uses a graph sharding algorithm on a proxy representation of the model. Merak also presents the non-intrusive API for scaling out foundation model training with minimal code modification. In addition, we design a high-performance 3D parallel runtime engine in Merak. It uses several techniques to exploit available training resources, including shifted critical path pipeline schedule that brings a higher computation utilization, stage-aware recomputation that makes use of idle worker memory, and sub-pipelined tensor model parallelism that overlaps communication and computation. Experiments on 64 GPUs show Merak can speedup the training performance over the state-of-the-art 3D parallelism frameworks of models with 1.5, 2.5, 8.3, and 20 billion parameters by up to 1.42X, 1.39X, 1.43X, and 1.61X, respectively.

cs.LG↗

arXiv2021

EmbRace: Accelerating Sparse Communication for Distributed Training of NLP Neural Networks

Distributed data-parallel training has been widely adopted for deep neural network (DNN) models. Although current deep learning (DL) frameworks scale well for dense models like image classification models, we find that these DL frameworks have relatively low scalability for sparse models like natural language processing (NLP) models that have highly sparse embedding tables. Most existing works overlook the sparsity of model parameters thus suffering from significant but unnecessary communication overhead. In this paper, we propose EmbRace, an efficient communication framework to accelerate communications of distributed training for sparse models. EmbRace introduces Sparsity-aware Hybrid Communication, which integrates AlltoAll and model parallelism into data-parallel training, so as to reduce the communication overhead of highly sparse parameters. To effectively overlap sparse communication with both backward and forward computation, EmbRace further designs a 2D Communication Scheduling approach which optimizes the model computation procedure, relaxes the dependency of embeddings, and schedules the sparse communications of each embedding row with a priority queue. We have implemented a prototype of EmbRace based on PyTorch and Horovod, and conducted comprehensive evaluations with four representative NLP models. Experimental results show that EmbRace achieves up to 2.41X speedup compared to the state-of-the-art distributed training baselines.

cs.LG↗

arXiv2016

A Parallel Solution to Finding Nodal Neighbors in Generic Meshes

In this paper we specifically present a parallel solution to finding the one-ring neighboring nodes and elements for each vertex in generic meshes. The finding of nodal neighbors is computationally straightforward but expensive for large meshes. To improve the efficiency, the parallelism is adopted by utilizing the modern Graphics Processing Unit (GPU). The presented parallel solution is heavily dependent on the parallel sorting, scan, and reduction, and can be applied to determine both the neighboring nodes and elements. To evaluate the performance, the parallel solution is compared to the corresponding serial solution. Experimental results show that: our parallel solution can achieve the speedups of approximately 55 and 90 over the corresponding serial solution for finding neighboring nodes and elements, respectively. Our parallel solution is efficient and easy to implement, but requires the allocation of large device memory.

cs.CG↗