TAPAS: Fast and Automatic Derivation of Tensor Parallel Strategies for Large Neural Networks
- URL: http://arxiv.org/abs/2302.00247v2
- Date: Tue, 05 Aug 2025 04:55:48 GMT
- Title: TAPAS: Fast and Automatic Derivation of Tensor Parallel Strategies for Large Neural Networks
- Authors: Ziji Shi, Le Jiang, Ang Wang, Jie Zhang, Chencan Wu, Yong Li, Xiaokui Xiao, Wei Lin, Jialin Li,
- Abstract summary: We build an automatic parallelism framework named TAPAS that eliminates redundant search efforts.<n> TAPAS employs a divide-and-linear approach that efficiently folds the search space by identifying those unique substructures.<n>Our evaluations demonstrate that TAPAS outperforms the state-of-the-art automatic parallelism frameworks by up to $160times$ in search speed.
- Score: 27.634123904734615
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Tensor parallelism is an essential technique for distributed training of large neural networks. However, automatically determining an optimal tensor parallel strategy is challenging due to the gigantic search space, which grows exponentially with model size and tensor dimension. This prohibits the adoption of auto-parallel systems on larger models. We observe that neural networks usually contain repeated substructures, and build an automatic parallelism framework named TAPAS that eliminates redundant search efforts. TAPAS employs a divide-and-conquer approach that efficiently folds the search space by identifying those unique substructures. As a result, it runs at sub-linear complexity concerning the model size, making it a scalable solution for training large-scale networks. Our evaluations demonstrate that TAPAS outperforms the state-of-the-art automatic parallelism frameworks by up to $160\times$ in search speed on a wide range of models, and the performance of derived strategies is competitive or even better compared with the expert-engineered Megatron-LM library.
Related papers
- Automatic Operator-level Parallelism Planning for Distributed Deep Learning -- A Mixed-Integer Programming Approach [6.449961842220686]
We propose a bi-level solution framework balancing optimality with computational efficiency.<n>Our framework achieves comparable or superior performance, reducing computational bubbles by half under the same memory constraints.<n>Such capabilities position our solution as both a valuable research tool for exploring optimal parallelization strategies and a practical industrial solution for large-scale AI deployment.
arXiv Detail & Related papers (2025-03-12T13:00:29Z) - Prediction-Assisted Online Distributed Deep Learning Workload Scheduling in GPU Clusters [24.845122459974466]
This paper proposes an adaptive shortest-remaining-processing-time-first (A-SRPT) scheduling algorithm.
By modeling each job as a graph corresponding to heterogeneous Deep Neural Network (DNN) models, A-SRPT strategically assigns jobs to the available GPU.
A-SRPT maps the complex scheduling problem into a single-machine instance, which is addressed optimally by a preemptive "shortest-remaining-processing-time-first" strategy.
arXiv Detail & Related papers (2025-01-09T20:19:01Z) - Universal Checkpointing: A Flexible and Efficient Distributed Checkpointing System for Large-Scale DNN Training with Reconfigurable Parallelis [16.04816181826873]
Universal Checkpointing (UCP) is a novel checkpointing system for deep neural network (DNN) training.<n>UCP overcomes challenges in existing systems by decoupling checkpoint structure from parallel training strategies and hardware configurations.<n>We present a pattern-based reconfiguration pipeline that enables automatic, flexible, and efficient mapping of checkpoint state to various parallelism strategies.
arXiv Detail & Related papers (2024-06-27T01:28:30Z) - OTOv3: Automatic Architecture-Agnostic Neural Network Training and
Compression from Structured Pruning to Erasing Operators [57.145175475579315]
This topic spans various techniques, from structured pruning to neural architecture search, encompassing both pruning and erasing operators perspectives.
We introduce the third-generation Only-Train-Once (OTOv3), which first automatically trains and compresses a general DNN through pruning and erasing operations.
Our empirical results demonstrate the efficacy of OTOv3 across various benchmarks in structured pruning and neural architecture search.
arXiv Detail & Related papers (2023-12-15T00:22:55Z) - Testing RadiX-Nets: Advances in Viable Sparse Topologies [0.9555447998395205]
Sparsification of hyper-parametrized deep neural networks (DNNs) creates simpler representations of complex data.
RadiX-Nets, a subgroup of DNNs, maintain runtime which counteracts their lack of neural connections.
This paper presents a testing suite for RadiX-Nets in scalable models.
arXiv Detail & Related papers (2023-11-06T23:27:28Z) - Entropic Score metric: Decoupling Topology and Size in Training-free NAS [18.804303642485895]
This paper contributes with a novel training-free metric, named Entropic Score, to estimate model expressivity through the aggregated element-wise entropy of its activations.
A proper combination with LogSynflow, to search for model size, yields superior capability to completely design high-performance Hybrid Transformers for edge applications in less than 1 GPU hour.
arXiv Detail & Related papers (2023-10-06T11:49:21Z) - T-GAE: Transferable Graph Autoencoder for Network Alignment [79.89704126746204]
T-GAE is a graph autoencoder framework that leverages transferability and stability of GNNs to achieve efficient network alignment without retraining.
Our experiments demonstrate that T-GAE outperforms the state-of-the-art optimization method and the best GNN approach by up to 38.7% and 50.8%, respectively.
arXiv Detail & Related papers (2023-10-05T02:58:29Z) - How neural networks learn to classify chaotic time series [77.34726150561087]
We study the inner workings of neural networks trained to classify regular-versus-chaotic time series.
We find that the relation between input periodicity and activation periodicity is key for the performance of LKCNN models.
arXiv Detail & Related papers (2023-06-04T08:53:27Z) - SWARM Parallelism: Training Large Models Can Be Surprisingly
Communication-Efficient [69.61083127540776]
Deep learning applications benefit from using large models with billions of parameters.
Training these models is notoriously expensive due to the need for specialized HPC clusters.
We consider alternative setups for training large models: using cheap "preemptible" instances or pooling existing resources from multiple regions.
arXiv Detail & Related papers (2023-01-27T18:55:19Z) - HARL: Hierarchical Adaptive Reinforcement Learning Based Auto Scheduler
for Neural Networks [51.71682428015139]
We propose HARL, a reinforcement learning-based auto-scheduler for efficient tensor program exploration.
HarL improves the tensor operator performance by 22% and the search speed by 4.3x compared to the state-of-the-art auto-scheduler.
Inference performance and search speed are also significantly improved on end-to-end neural networks.
arXiv Detail & Related papers (2022-11-21T04:15:27Z) - Intelligence Processing Units Accelerate Neuromorphic Learning [52.952192990802345]
Spiking neural networks (SNNs) have achieved orders of magnitude improvement in terms of energy consumption and latency.
We present an IPU-optimized release of our custom SNN Python package, snnTorch.
arXiv Detail & Related papers (2022-11-19T15:44:08Z) - On Optimizing the Communication of Model Parallelism [74.15423270435949]
We study a novel and important communication pattern in large-scale model-parallel deep learning (DL)
In cross-mesh resharding, a sharded tensor needs to be sent from a source device mesh to a destination device mesh.
We propose two contributions to address cross-mesh resharding: an efficient broadcast-based communication system, and an "overlapping-friendly" pipeline schedule.
arXiv Detail & Related papers (2022-11-10T03:56:48Z) - DistIR: An Intermediate Representation and Simulator for Efficient
Neural Network Distribution [15.086401550425125]
DistIR is a representation for distributed computation that is tailored for efficient analyses.
We show how DistIR and its simulator enable fast grid searches over complex distribution spaces spanning up to 1000+ configurations.
arXiv Detail & Related papers (2021-11-09T21:32:51Z) - A quantum algorithm for training wide and deep classical neural networks [72.2614468437919]
We show that conditions amenable to classical trainability via gradient descent coincide with those necessary for efficiently solving quantum linear systems.
We numerically demonstrate that the MNIST image dataset satisfies such conditions.
We provide empirical evidence for $O(log n)$ training of a convolutional neural network with pooling.
arXiv Detail & Related papers (2021-07-19T23:41:03Z) - Mitigating Performance Saturation in Neural Marked Point Processes:
Architectures and Loss Functions [50.674773358075015]
We propose a simple graph-based network structure called GCHP, which utilizes only graph convolutional layers.
We show that GCHP can significantly reduce training time and the likelihood ratio loss with interarrival time probability assumptions can greatly improve the model performance.
arXiv Detail & Related papers (2021-07-07T16:59:14Z) - Parareal Neural Networks Emulating a Parallel-in-time Algorithm [1.988145627448243]
As deep neural networks (DNNs) become deeper, the training time increases.
In this paper, we introduce a novel methodology to construct a parallel neural network.
arXiv Detail & Related papers (2021-03-16T02:03:39Z) - Parallel Training of Deep Networks with Local Updates [84.30918922367442]
Local parallelism is a framework which parallelizes training of individual layers in deep networks by replacing global backpropagation with truncated layer-wise backpropagation.
We show results in both vision and language domains across a diverse set of architectures, and find that local parallelism is particularly effective in the high-compute regime.
arXiv Detail & Related papers (2020-12-07T16:38:45Z) - Auto-MAP: A DQN Framework for Exploring Distributed Execution Plans for
DNN Workloads [11.646744408920764]
Auto-MAP is a framework for exploring distributed execution plans for workloads.
It can automatically discovering fast parallelization strategies through reinforcement learning on IR level of deep learning models.
Our evaluation shows that Auto-MAP can find the optimal solution in two hours, while achieving better throughput on several NLP and convolution models.
arXiv Detail & Related papers (2020-07-08T12:38:03Z) - Auto-PyTorch Tabular: Multi-Fidelity MetaLearning for Efficient and
Robust AutoDL [53.40030379661183]
Auto-PyTorch is a framework to enable fully automated deep learning (AutoDL)
It combines multi-fidelity optimization with portfolio construction for warmstarting and ensembling of deep neural networks (DNNs)
We show that Auto-PyTorch performs better than several state-of-the-art competitors on average.
arXiv Detail & Related papers (2020-06-24T15:15:17Z) - A Linear Algebraic Approach to Model Parallelism in Deep Learning [0.0]
Training deep neural networks (DNNs) in large-cluster computing environments is increasingly necessary, as networks grow in size and complexity.
We propose a linear-algebraic approach to model parallelism in deep learning, which allows parallel distribution of any tensor in the DNN.
We build distributed DNN layers using these parallel primitives, composed with sequential layer implementations, and demonstrate their application by building and training a distributed DNN using DistDL, a PyTorch and MPI-based distributed deep learning toolkit.
arXiv Detail & Related papers (2020-06-04T19:38:05Z) - Large-Scale Gradient-Free Deep Learning with Recursive Local
Representation Alignment [84.57874289554839]
Training deep neural networks on large-scale datasets requires significant hardware resources.
Backpropagation, the workhorse for training these networks, is an inherently sequential process that is difficult to parallelize.
We propose a neuro-biologically-plausible alternative to backprop that can be used to train deep networks.
arXiv Detail & Related papers (2020-02-10T16:20:02Z)
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.