MAD Max Beyond Single-Node: Enabling Large Machine Learning Model Acceleration on Distributed Systems

• URL: http://arxiv.org/abs/2310.02784v3
• Date: Mon, 10 Jun 2024 20:31:07 GMT
• Title: MAD Max Beyond Single-Node: Enabling Large Machine Learning Model Acceleration on Distributed Systems
• Authors: Samuel Hsia, Alicia Golden, Bilge Acun, Newsha Ardalani, Zachary DeVito, Gu-Yeon Wei, David Brooks, Carole-Jean Wu,
• Abstract summary: Training and deploying large-scale machine learning models is time-consuming, requires significant distributed computing infrastructures, and incurs high operational costs. To minimize this outstanding communication latency and other inherent at-scale inefficiencies, we introduce an agile performance modeling framework, MAD-Max. This framework is designed to optimize parallelization strategies and facilitate hardware-software co-design opportunities.
• Score: 6.8519529064678375
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 datacenter-scale infrastructures, reveals that 14~32% of all GPU hours are spent on communication with no overlapping computation. To minimize this outstanding communication latency and other inherent at-scale inefficiencies, we introduce an agile performance modeling framework, MAD-Max. This framework is designed to optimize parallelization strategies and facilitate hardware-software co-design opportunities. Through the application of MAD-Max to a suite of real-world large-scale ML models on state-of-the-art GPU clusters, we showcase potential throughput enhancements of up to 2.24x for pre-training and up to 5.2x for inference scenarios, respectively.

Related papers

• Parm: Efficient Training of Large Sparsely-Activated Models with Dedicated Schedules [15.680276212483292]
  We propose Parm, a system that accelerates MP+EP+ESP training by designing two dedicated schedules for placing communication tasks. Parm achieves 1.13$times$ to 5.77$times$ speedup on 1296 manually configured MoE layers and approximately 3$times$ improvement on two real-world MoE models.
  arXiv  Detail & Related papers  (2024-06-30T05:55:11Z)
• Towards Universal Performance Modeling for Machine Learning Training on Multi-GPU Platforms [4.959530958049395]
  We develop a pipeline to Characterize and predict the training performance of modern machine learning (ML) workloads on compute systems. Our pipeline generalizes to other types of ML workloads, such as Transformer-based NLP models. It is capable of generating insights such as quickly selecting the fastest embedding table sharding configuration.
  arXiv  Detail & Related papers  (2024-04-19T07:20:33Z)
• Analysis of Distributed Optimization Algorithms on a Real Processing-In-Memory System [21.09681871279162]
  Processing-In-Memory (PIM) is a promising solution to alleviate the data movement bottleneck. Our goal is to understand the capabilities and characteristics of popular distributed optimization algorithms on real-world PIM architectures.
  arXiv  Detail & Related papers  (2024-04-10T17:00:04Z)
• Federated Fine-Tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly [62.473245910234304]
  This paper takes a hardware-centric approach to explore how Large Language Models can be brought to modern edge computing systems. We provide a micro-level hardware benchmark, compare the model FLOP utilization to a state-of-the-art data center GPU, and study the network utilization in realistic conditions.
  arXiv  Detail & Related papers  (2023-10-04T20:27:20Z)
• In Situ Framework for Coupling Simulation and Machine Learning with Application to CFD [51.04126395480625]
  Recent years have seen many successful applications of machine learning (ML) to facilitate fluid dynamic computations. As simulations grow, generating new training datasets for traditional offline learning creates I/O and storage bottlenecks. This work offers a solution by simplifying this coupling and enabling in situ training and inference on heterogeneous clusters.
  arXiv  Detail & Related papers  (2023-06-22T14:07:54Z)
• FlexMoE: Scaling Large-scale Sparse Pre-trained Model Training via Dynamic Device Placement [19.639936387834677]
  Mixture-of-Experts (MoEs) are becoming more popular and have demonstrated impressive pretraining scalability in various downstream tasks. MoEs are becoming a new data analytics paradigm in the data life cycle and suffering from unique challenges at scales, complexities, and granularities never before possible. In this paper, we propose a novel DNN training framework, FlexMoE, which systematically and transparently address the inefficiency caused by dynamic dataflow.
  arXiv  Detail & Related papers  (2023-04-08T07:34:26Z)
• Partitioning Distributed Compute Jobs with Reinforcement Learning and Graph Neural Networks [58.720142291102135]
  Large-scale machine learning models are bringing advances to a broad range of fields. Many of these models are too large to be trained on a single machine, and must be distributed across multiple devices. We show that maximum parallelisation is sub-optimal in relation to user-critical metrics such as throughput and blocking rate.
  arXiv  Detail & Related papers  (2023-01-31T17:41:07Z)
• Large Batch Simulation for Deep Reinforcement Learning [101.01408262583378]
  We accelerate deep reinforcement learning-based training in visually complex 3D environments by two orders of magnitude over prior work. We realize end-to-end training speeds of over 19,000 frames of experience per second on a single and up to 72,000 frames per second on a single eight- GPU machine. By combining batch simulation and performance optimizations, we demonstrate that Point navigation agents can be trained in complex 3D environments on a single GPU in 1.5 days to 97% of the accuracy of agents trained on a prior state-of-the-art system.
  arXiv  Detail & Related papers  (2021-03-12T00:22:50Z)
• Edge-assisted Democratized Learning Towards Federated Analytics [67.44078999945722]
  We show the hierarchical learning structure of the proposed edge-assisted democratized learning mechanism, namely Edge-DemLearn. We also validate Edge-DemLearn as a flexible model training mechanism to build a distributed control and aggregation methodology in regions.
  arXiv  Detail & Related papers  (2020-12-01T11:46:03Z)
• A Survey on Large-scale Machine Learning [67.6997613600942]
  Machine learning can provide deep insights into data, allowing machines to make high-quality predictions. Most sophisticated machine learning approaches suffer from huge time costs when operating on large-scale data. Large-scale Machine Learning aims to learn patterns from big data with comparable performance efficiently.
  arXiv  Detail & Related papers  (2020-08-10T06:07:52Z)
• Deep Generative Models that Solve PDEs: Distributed Computing for Training Large Data-Free Models [25.33147292369218]
  Recent progress in scientific machine learning (SciML) has opened up the possibility of training novel neural network architectures that solve complex partial differential equations (PDEs) Here we report on a software framework for data parallel distributed deep learning that resolves the twin challenges of training these large SciML models. Our framework provides several out of the box functionality including (a) loss integrity independent of number of processes, (b) synchronized batch normalization, and (c) distributed higher-order optimization methods.
  arXiv  Detail & Related papers  (2020-07-24T22:42:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.