Aishwarya Sarkar, Sayan Ghosh, Nathan Tallent et al.
Large-scale Graph Neural Networks (GNNs) are typically trained by sampling a vertex's neighbors to a fixed distance. Because large input graphs are distributed, training requires frequent irregular communication that stalls forward progress. Moreover, fetched data changes with graph, graph distribution, sample and batch parameters, and caching polices. Consequently, any static prefetching method will miss crucial opportunities to adapt to different dynamic conditions. In this paper, we introduce Rudder, a software module embedded in the state-of-the-art AWS DistDGL framework, to autonomously prefetch remote nodes and minimize communication. Rudder's adaptation contrasts with both standard heuristics and traditional ML classifiers. We observe that the generative AI found in contemporary Large Language Models (LLMs) exhibits emergent properties like In-Context Learning (ICL) for zero-shot tasks, with logical multi-step reasoning. We find this behavior well-suited for adaptive control even with substantial undertraining. Evaluations using standard datasets and unseen configurations on the NERSC Perlmutter supercomputer show up to 91% improvement in end-to-end training performance over baseline DistDGL (no prefetching), and an 82% improvement over static prefetching, reducing communication by over 50%. Our code is available at https://github.com/aishwaryyasarkar/rudder-llm-agent.
Cunyang Wei, Siddharth Singh, Aishwarya Sarkar et al.
Graph neural networks (GNNs) are widely used for learning on graph datasets derived from various real-world scenarios. Learning from extremely large graphs requires distributed training, and mini-batching with sampling is a popular approach for parallelizing GNN training. Existing distributed mini-batch approaches have significant performance bottlenecks due to expensive sampling methods and limited scaling when using data parallelism. In this work, we present ScaleGNN, a 4D parallel framework for scalable mini-batch GNN training that combines communication-free distributed sampling, 3D parallel matrix multiplication (PMM), and data parallelism. ScaleGNN introduces a uniform vertex sampling algorithm, enabling each process (GPU device) to construct its local mini-batch, i.e., subgraph partitions without any inter-process communication. 3D PMM enables scaling mini-batch training to much larger GPU counts than vanilla data parallelism with significantly lower communication overheads. We also present additional optimizations to overlap sampling with training, reduce communication overhead by sending data in lower precision, kernel fusion, and communication-computation overlap. We evaluate ScaleGNN on five graph datasets and demonstrate strong scaling up to 2048 GPUs on Perlmutter, 2048 GCDs on Frontier, and 1024 GPUs on Tuolumne. On Perlmutter, ScaleGNN achieves 3.5x end-to-end training speedup over the SOTA baseline on ogbn-products.
Searching for the $k$-nearest neighbors (KNN) in multimodal data retrieval is computationally expensive, particularly due to the inherent difficulty in comparing similarity measures across different modalities. Recent advances in multimodal machine learning address this issue by mapping data into a shared embedding space; however, the high dimensionality of these embeddings (hundreds to thousands of dimensions) presents a challenge for time-sensitive vision applications. This work proposes Order-Preserving Dimension Reduction (OPDR), aiming to reduce the dimensionality of embeddings while preserving the ranking of KNN in the lower-dimensional space. One notable component of OPDR is a new measure function to quantify KNN quality as a global metric, based on which we derive a closed-form map between target dimensionality and key contextual parameters. We have integrated OPDR with multiple state-of-the-art dimension-reduction techniques, distance functions, and embedding models; experiments on a variety of multimodal datasets demonstrate that OPDR effectively retains recall high accuracy while significantly reducing computational costs.
Aishwarya Sarkar, Sayan Ghosh, Nathan R. Tallent et al.
Successful machine learning on graphs or networks requires embeddings that not only represent nodes and edges as low-dimensional vectors but also preserve the graph structure. Established methods for generating embeddings require flexible exploration of the entire graph through repeated use of random walks that capture graph structure with samples of nodes and edges. These methods create scalability challenges for massive graphs with millions-to-billions of edges because single-node solutions have inadequate memory and processing capabilities. We present NOMAD, a distributed-memory graph embedding framework using the Message Passing Interface (MPI) for distributed graphs. NOMAD implements proximity-based models proposed in the widely popular LINE (Large-scale Information Network Embedding) algorithm. We propose several practical trade-offs to improve the scalability and communication overheads confronted by irregular and distributed graph embedding methods, catering to massive-scale graphs arising in web and science domains. NOMAD demonstrates median speedups of 10/100x on CPU-based NERSC Perlmutter cluster relative to the popular reference implementations of multi-threaded LINE and node2vec, 35-76x over distributed PBG, and competitive embedding quality relative to LINE, node2vec, and GraphVite, while yielding 12-370x end-to-end speedups on real-world graphs.
Image segmentation is a critical enabler for tasks ranging from medical diagnostics to autonomous driving. However, the correct segmentation semantics - where are boundaries located? what segments are logically similar? - change depending on the domain, such that state-of-the-art foundation models can generate meaningless and incorrect results. Moreover, in certain domains, fine-tuning and retraining techniques are infeasible: obtaining labels is costly and time-consuming; domain images (micrographs) can be exponentially diverse; and data sharing (for third-party retraining) is restricted. To enable rapid adaptation of the best segmentation technology, we propose the concept of semantic boosting: given a zero-shot foundation model, guide its segmentation and adjust results to match domain expectations. We apply semantic boosting to the Segment Anything Model (SAM) to obtain microstructure segmentation for transmission electron microscopy. Our booster, SAM-I-Am, extracts geometric and textural features of various intermediate masks to perform mask removal and mask merging operations. We demonstrate a zero-shot performance increase of (absolute) +21.35%, +12.6%, +5.27% in mean IoU, and a -9.91%, -18.42%, -4.06% drop in mean false positive masks across images of three difficulty classes over vanilla SAM (ViT-L).
Neural Architecture Search (NAS) is a powerful approach of automating the design of efficient neural architectures. In contrast to traditional NAS methods, recently proposed one-shot NAS methods prove to be more efficient in performing NAS. One-shot NAS works by generating a singular weight-sharing supernetwork that acts as a search space (container) of subnetworks. Despite its achievements, designing the one-shot search space remains a major challenge. In this work we propose a search space design strategy for Vision Transformer (ViT)-based architectures. In particular, we convert the Segment Anything Model (SAM) into a weight-sharing supernetwork called SuperSAM. Our approach involves automating the search space design via layer-wise structured pruning and parameter prioritization. While the structured pruning applies probabilistic removal of certain transformer layers, parameter prioritization performs weight reordering and slicing of MLP-blocks in the remaining layers. We train supernetworks on several datasets using the sandwich rule. For deployment, we enhance subnetwork discovery by utilizing a program autotuner to identify efficient subnetworks within the search space. The resulting subnetworks are 30-70% smaller in size compared to the original pre-trained SAM ViT-B, yet outperform the pretrained model. Our work introduces a new and effective method for ViT NAS search-space design.
Aishwarya Sarkar, Sayan Ghosh, Nathan R. Tallent et al.
Graph Neural Networks (GNN) are indispensable in learning from graph-structured data, yet their rising computational costs, especially on massively connected graphs, pose significant challenges in terms of execution performance. To tackle this, distributed-memory solutions such as partitioning the graph to concurrently train multiple replicas of GNNs are in practice. However, approaches requiring a partitioned graph usually suffer from communication overhead and load imbalance, even under optimal partitioning and communication strategies due to irregularities in the neighborhood minibatch sampling. This paper proposes practical trade-offs for improving the sampling and communication overheads for representation learning on distributed graphs (using popular GraphSAGE architecture) by developing a parameterized continuous prefetch and eviction scheme on top of the state-of-the-art Amazon DistDGL distributed GNN framework, demonstrating about 15-40% improvement in end-to-end training performance on the National Energy Research Scientific Computing Center's (NERSC) Perlmutter supercomputer for various OGB datasets.
Automating the theory-experiment cycle requires effective distributed workflows that utilize a computing continuum spanning lab instruments, edge sensors, computing resources at multiple facilities, data sets distributed across multiple information sources, and potentially cloud. Unfortunately, the obvious methods for constructing continuum platforms, orchestrating workflow tasks, and curating datasets over time fail to achieve scientific requirements for performance, energy, security, and reliability. Furthermore, achieving the best use of continuum resources depends upon the efficient composition and execution of workflow tasks, i.e., combinations of numerical solvers, data analytics, and machine learning. Pacific Northwest National Laboratory's LDRD "Cloud, High-Performance Computing (HPC), and Edge for Science and Security" (CHESS) has developed a set of interrelated capabilities for enabling distributed scientific workflows and curating datasets. This report describes the results and successes of CHESS from the perspective of open science.
Md Anwar Hossen, Nathan R. Tallent, Luanzheng Guo et al.
Scientific discovery increasingly requires learning on federated datasets, fed by streams from high-resolution instruments, that have extreme class imbalance. Current ML approaches either require impractical data aggregation or fail due to class imbalance. Existing coreset selection methods rely on local heuristics, making them unaware of the global data landscape and prone to sub-optimal and non-representative pruning. To overcome these challenges, we introduce SCOPE (Semantic Coreset using Orthogonal Projection Embeddings for Federated learning), a coreset framework for federated data that filters anomalies and adaptively prunes redundant data to mitigate long-tail skew. By analyzing the latent space distribution, we score each data point using a representation score that measures the reliability of core class features, a diversity score that quantifies the novelty of orthogonal residuals, and a boundary proximity score that indicates similarity to competing classes. Unlike prior methods, SCOPE shares only scalar metrics with a federated server to construct a global consensus, ensuring communication efficiency. Guided by the global consensus, SCOPE dynamically filters local noise and discards redundant samples to counteract global feature skews. Extensive experiments demonstrate that SCOPE yields competitive global accuracy and robust convergence, all while achieving exceptional efficiency with a 128x to 512x reduction in uplink bandwidth, a 7.72x wall-clock acceleration and reduced FLOP and VRAM footprints for local coreset selection.