Neural Information Processing Systems http://nips.cc/

To improve the resilience of distributed training to worst-case, or Byzantine node failures, several recent methods have replaced gradient averaging with robust aggregation methods. Such techniques can have high computational costs, often quadratic in the number of compute nodes, and only have limited robustness guarantees. Other methods have instead used redundancy to guarantee robustness, but can only tolerate limited numbers of Byzantine failures. In this work, we present DETOX, a Byzantine-resilient distributed training framework that combines algorithmic redundancy with robust aggregation. DETOX operates in two steps, a filtering step that uses limited redundancy to significantly reduce the effect of Byzantine nodes, and a hierarchical aggregation step that can be used in tandem with any state-of-the-art robust aggregation method. We show theoretically that this leads to a substantial increase in robustness, and has a per iteration runtime that can be nearly linear in the number of compute nodes. We provide extensive experiments over real distributed setups across a variety of large-scale machine learning tasks, showing that DETOX leads to orders of magnitude accuracy and speedup improvements over many state-of-the-art Byzantine-resilient approaches.

Training deep neural networks on large datasets can often be accelerated by using multiple compute nodes. This approach, known as distributed training, can utilize hundreds of computers via specialized message-passing protocols such as Ring All-Reduce.However, running these protocols at scale requires reliable high-speed networking that is only available in dedicated clusters.In contrast, many real-world applications, such as federated learning and cloud-based distributed training, operate on unreliable devices with unstable network bandwidth.As a result, these applications are restricted to using parameter servers or gossip-based averaging protocols.In this work, we lift that restriction by proposing Moshpit All-Reduce -- an iterative averaging protocol that exponentially converges to the global average.We demonstrate the efficiency of our protocol for distributed optimization with strong theoretical guarantees.The experiments show 1.3x speedup for ResNet-50 training on ImageNet compared to competitive gossip-based strategies and 1.5x speedup when training ALBERT-large on preemptible compute nodes.

Neural Information Processing Systems http://nips.cc/

Neural Information Processing Systems http://nips.cc/

Network failure diagnosis is challenging yet critical for high-performance computing (HPC) systems. Existing methods cannot be directly applied to HPC scenarios due to data heterogeneity and lack of accuracy. This paper proposes a novel framework, called ClusterRCA, to localize culprit nodes and determine failure types by leveraging multimodal data. ClusterRCA extracts features from topologically connected network interface controller (NIC) pairs to analyze the diverse, multimodal data in HPC systems. To accurately localize culprit nodes and determine failure types, ClusterRCA combines classifier-based and graph-based approaches. A failure graph is constructed based on the output of the state classifier, and then it performs a customized random walk on the graph to localize the root cause. Experiments on datasets collected by a top-tier global HPC device vendor show ClusterRCA achieves high accuracy in diagnosing network failure for HPC systems. ClusterRCA also maintains robust performance across different application scenarios.

Compute nodes on modern heterogeneous supercomputing systems comprise CPUs, GPUs, and high-speed network interconnects (NICs). Parallelization is identified as a technique for effectively utilizing these systems to execute scalable simulation and deep learning workloads. The resulting inter-process communication from the distributed execution of these parallel workloads is one of the key factors contributing to its performance bottleneck. Most programming models and runtime systems enabling the communication requirements on these systems support GPU-aware communication schemes that move the GPU-attached communication buffers in the application directly from the GPU to the NIC without staging through the host memory. A CPU thread is required to orchestrate the communication operations even with support for such GPU-awareness. This survey discusses various available GPU-centric communication schemes that move the control path of the communication operations from the CPU to the GPU. This work presents the need for the new communication schemes, various GPU and NIC capabilities required to implement the schemes, and the potential use-cases addressed. Based on these discussions, challenges involved in supporting the exhibited GPU-centric communication schemes are discussed.

Training deep neural networks on large datasets can often be accelerated by using multiple compute nodes. This approach, known as distributed training, can utilize hundreds of computers via specialized message-passing protocols such as Ring All-Reduce.However, running these protocols at scale requires reliable high-speed networking that is only available in dedicated clusters.In contrast, many real-world applications, such as federated learning and cloud-based distributed training, operate on unreliable devices with unstable network bandwidth.As a result, these applications are restricted to using parameter servers or gossip-based averaging protocols.In this work, we lift that restriction by proposing Moshpit All-Reduce -- an iterative averaging protocol that exponentially converges to the global average.We demonstrate the efficiency of our protocol for distributed optimization with strong theoretical guarantees.The experiments show 1.3x speedup for ResNet-50 training on ImageNet compared to competitive gossip-based strategies and 1.5x speedup when training ALBERT-large on preemptible compute nodes.

Increasing HPC cluster sizes and large-scale simulations that produce petabytes of data per run, create massive IO and storage challenges for analysis. Deep learning-based techniques, in particular, make use of these amounts of domain data to extract patterns that help build scientific understanding. Here, we demonstrate a streaming workflow in which simulation data is streamed directly to a machine-learning (ML) framework, circumventing the file system bottleneck. Data is transformed in transit, asynchronously to the simulation and the training of the model. With the presented workflow, data operations can be performed in common and easy-to-use programming languages, freeing the application user from adapting the application output routines. As a proof-of-concept we consider a GPU accelerated particle-in-cell (PIConGPU) simulation of the Kelvin- Helmholtz instability (KHI). We employ experience replay to avoid catastrophic forgetting in learning from this non-steady process in a continual manner. We detail challenges addressed while porting and scaling to Frontier exascale system.