Related papers: LuWu: An End-to-End In-Network Out-of-Core Optimiz…

To train modern large DNN models, pipeline parallelism has recently emerged, which distributes the model across GPUs and enables different devices to process different microbatches in pipeline. Earlier pipeline designs allow multiple…

Large deep learning models have shown great potential for delivering exceptional results in various applications. However, the training process can be incredibly challenging due to the models' vast parameter sizes, often consisting of…

Training and deploying large-scale machine learning models is time-consuming, requires significant distributed computing infrastructures, and incurs high operational costs. Our analysis, grounded in real-world large model training on…

The dedicated memory of hardware accelerators can be insufficient to store all weights and/or intermediate states of large deep learning models. Although model parallelism is a viable approach to reduce the memory pressure issue,…

Scaling up model depth and size is now a common approach to raise accuracy in many deep learning (DL) applications, as evidenced by the widespread success of multi-billion or even trillion parameter models in natural language processing…

With the rapid adoption of Large Language Models (LLMs), LLM-adapters have become increasingly common, providing lightweight specialization of large-scale models. Serving hundreds or thousands of these adapters on a single GPU allows…

Large-scale Pretrained Language Models (PLMs) have become the new paradigm for Natural Language Processing (NLP). PLMs with hundreds of billions parameters such as GPT-3 have demonstrated strong performances on natural language…

Reinforcement Learning (RL) has become the most effective post-training approach for improving the capabilities of Large Language Models (LLMs). In practice, because of the high demands on latency and memory, it is particularly challenging…

Overlapping communication with computation is crucial for distributed large-model training, yet optimizing it - especially when computation becomes the bottleneck-remains challenging. We present Lagom, a system that co-tunes communication…

The increasing complexity of deep learning recommendation models (DLRM) has led to a growing need for large-scale distributed systems that can efficiently train vast amounts of data. In DLRM, the sparse embedding table is a crucial…

Benefitting from large-scale training datasets and the complex training network, Convolutional Neural Networks (CNNs) are widely applied in various fields with high accuracy. However, the training process of CNNs is very time-consuming,…

Training a Graph Neural Network (GNN) model on large-scale graphs involves a high volume of data communication and computations. While state-of-the-art CPUs and GPUs feature high computing power, the Standard GNN training protocol adopted…

Training large language models (LLMs) is fundamentally constrained by limited device memory and costly inter-device communication. Although pipeline parallelism alleviates memory pressure by partitioning models across devices, it incurs…

Deep learning models are trained on servers with many GPUs, and training must scale with the number of GPUs. Systems such as TensorFlow and Caffe2 train models with parallel synchronous stochastic gradient descent: they process a batch of…

Large language models (LLMs) require enormous computing power to pretrain on massive datasets. When limited datasets are available, smaller-sized LLMs are better choice to pretrain (on user-specified datasets) by following the scaling laws…

The employment of high-performance servers and GPU accelerators for training deep neural network models have greatly accelerated recent advances in deep learning (DL). DL frameworks, such as TensorFlow, MXNet, and Caffe2, have emerged to…

Large Language Models (LLMs) built on transformer architectures have transformed natural language processing, achieving remarkable performance across diverse applications. While distributed inference frameworks enable practical deployment…

The proliferation of extensive neural network architectures, particularly deep learning models, presents a challenge in terms of resource-intensive training. GPU memory constraints have become a notable bottleneck in training such sizable…

We introduce Breadth-First Pipeline Parallelism, a novel training schedule which optimizes the combination of pipeline and data parallelism. Breadth-First Pipeline Parallelism lowers training time, cost and memory usage by combining a high…

This paper presents a low-cost network architecture for training large language models (LLMs) at hyperscale. We study the optimal parallelization strategy of LLMs and propose a novel datacenter network design tailored to LLM's unique…