Jaeyong Song

Jaeyong Song, Hongsun Jang, Jaewon Jung et al.

Graph neural networks (GNNs) are one of the rapidly growing fields within deep learning. While many distributed GNN training frameworks have been proposed to increase the training throughput, they face three limitations when applied to multi-server clusters. 1) They suffer from an inter-server communication bottleneck because they do not consider the inter-/intra-server bandwidth gap, a representative characteristic of multi-server clusters. 2) Redundant memory usage and computation hinder the scalability of the distributed frameworks. 3) Sampling methods, de facto standard in mini-batch training, incur unnecessary errors in multi-server clusters. We found that these limitations can be addressed by exploiting the characteristics of multi-server clusters. Here, we propose GraNNDis, a fast distributed GNN training framework for multi-server clusters. Firstly, we present Flexible Preloading, which preloads the essential vertex dependencies server-wise to reduce the low-bandwidth inter-server communications. Secondly, we introduce Cooperative Batching, which enables memory-efficient, less redundant mini-batch training by utilizing high-bandwidth intra-server communications. Thirdly, we propose Expansion-aware Sampling, a cluster-aware sampling method, which samples the edges that affect the system speedup. As sampling the intra-server dependencies does not contribute much to the speedup as they are communicated through fast intra-server links, it only targets a server boundary to be sampled. Lastly, we introduce One-Hop Graph Masking, a computation and communication structure to realize the above methods in multi-server environments. We evaluated GraNNDis on multi-server clusters, and it provided significant speedup over the state-of-the-art distributed GNN training frameworks. GraNNDis is open-sourced at https://github.com/AIS-SNU/GraNNDis_Artifact to facilitate its use.

Mingi Yoo, Jaeyong Song, Jounghoo Lee et al.

Graph convolutional networks (GCNs) are becoming increasingly popular as they overcome the limited applicability of prior neural networks. A GCN takes as input an arbitrarily structured graph and executes a series of layers which exploit the graph's structure to calculate their output features. One recent trend in GCNs is the use of deep network architectures. As opposed to the traditional GCNs which only span around two to five layers deep, modern GCNs now incorporate tens to hundreds of layers with the help of residual connections. From such deep GCNs, we find an important characteristic that they exhibit very high intermediate feature sparsity. We observe that with deep layers and residual connections, the number of zeros in the intermediate features sharply increases. This reveals a new opportunity for accelerators to exploit in GCN executions that was previously not present. In this paper, we propose SGCN, a fast and energy-efficient GCN accelerator which fully exploits the sparse intermediate features of modern GCNs. SGCN suggests several techniques to achieve significantly higher performance and energy efficiency than the existing accelerators. First, SGCN employs a GCN-friendly feature compression format. We focus on reducing the off-chip memory traffic, which often is the bottleneck for GCN executions. Second, we propose microarchitectures for seamlessly handling the compressed feature format. Third, to better handle locality in the existence of the varying sparsity, SGCN employs sparsity-aware cooperation. Sparsity-aware cooperation creates a pattern that exhibits multiple reuse windows, such that the cache can capture diverse sizes of working sets and therefore adapt to the varying level of sparsity. We show that SGCN achieves 1.71x speedup and 43.9% higher energy efficiency compared to the existing accelerators.

Jaeyong Song, Jinkyu Yim, Jaewon Jung et al.

In training of modern large natural language processing (NLP) models, it has become a common practice to split models using 3D parallelism to multiple GPUs. Such technique, however, suffers from a high overhead of inter-node communication. Compressing the communication is one way to mitigate the overhead by reducing the inter-node traffic volume; however, the existing compression techniques have critical limitations to be applied for NLP models with 3D parallelism in that 1) only the data parallelism traffic is targeted, and 2) the existing compression schemes already harm the model quality too much. In this paper, we present Optimus-CC, a fast and scalable distributed training framework for large NLP models with aggressive communication compression. Optimus-CC differs from existing communication compression frameworks in the following ways: First, we compress pipeline parallel (inter-stage) traffic. In specific, we compress the inter-stage backpropagation and the embedding synchronization in addition to the existing data-parallel traffic compression methods. Second, we propose techniques to avoid the model quality drop that comes from the compression. We further provide mathematical and empirical analyses to show that our techniques can successfully suppress the compression error. Lastly, we analyze the pipeline and opt to selectively compress those traffic lying on the critical path. This further helps reduce the compression error. We demonstrate our solution on a GPU cluster, and achieve superior speedup from the baseline state-of-the-art solutions for distributed training without sacrificing the model quality.

Mingi Yoo, Jaeyong Song, Hyeyoon Lee et al.

Graph convolutional networks (GCNs) are becoming increasingly popular as they can process a wide variety of data formats that prior deep neural networks cannot easily support. One key challenge in designing hardware accelerators for GCNs is the vast size and randomness in their data access patterns which greatly reduces the effectiveness of the limited on-chip cache. Aimed at improving the effectiveness of the cache by mitigating the irregular data accesses, prior studies often employ the vertex tiling techniques used in traditional graph processing applications. While being effective at enhancing the cache efficiency, those approaches are often sensitive to the tiling configurations where the optimal setting heavily depends on target input datasets. Furthermore, the existing solutions require manual tuning through trial-and-error or rely on sub-optimal analytical models. In this paper, we propose Slice-and-Forge (SnF), an efficient hardware accelerator for GCNs which greatly improves the effectiveness of the limited on-chip cache. SnF chooses a tiling strategy named feature slicing that splits the features into vertical slices and processes them in the outermost loop of the execution. This particular choice results in a repetition of the identical computational patterns over irregular graph data over multiple rounds. Taking advantage of such repetitions, SnF dynamically tunes its tile size. Our experimental results reveal that SnF can achieve 1.73x higher performance in geomean compared to prior work on multi-engine settings, and 1.46x higher performance in geomean on small scale settings, without the need for off-line analyses.

Hongsun Jang, Jaewon Jung, Jaeyong Song et al.

Training large deep neural network models is highly challenging due to their tremendous computational and memory requirements. Blockwise distillation provides one promising method towards faster convergence by splitting a large model into multiple smaller models. In state-of-the-art blockwise distillation methods, training is performed block-by-block in a data-parallel manner using multiple GPUs. To produce inputs for the student blocks, the teacher model is executed from the beginning until the current block under training. However, this results in a high overhead of redundant teacher execution, low GPU utilization, and extra data loading. To address these problems, we propose Pipe-BD, a novel parallelization method for blockwise distillation. Pipe-BD aggressively utilizes pipeline parallelism for blockwise distillation, eliminating redundant teacher block execution and increasing per-device batch size for better resource utilization. We also extend to hybrid parallelism for efficient workload balancing. As a result, Pipe-BD achieves significant acceleration without modifying the mathematical formulation of blockwise distillation. We implement Pipe-BD on PyTorch, and experiments reveal that Pipe-BD is effective on multiple scenarios, models, and datasets.

Hongsun Jang, Jaeyong Song, Jaewon Jung et al.

The recent huge advance of Large Language Models (LLMs) is mainly driven by the increase in the number of parameters. This has led to substantial memory capacity requirements, necessitating the use of dozens of GPUs just to meet the capacity. One popular solution to this is storage-offloaded training, which uses host memory and storage as an extended memory hierarchy. However, this obviously comes at the cost of storage bandwidth bottleneck because storage devices have orders of magnitude lower bandwidth compared to that of GPU device memories. Our work, Smart-Infinity, addresses the storage bandwidth bottleneck of storage-offloaded LLM training using near-storage processing devices on a real system. The main component of Smart-Infinity is SmartUpdate, which performs parameter updates on custom near-storage accelerators. We identify that moving parameter updates to the storage side removes most of the storage traffic. In addition, we propose an efficient data transfer handler structure to address the system integration issues for Smart-Infinity. The handler allows overlapping data transfers with fixed memory consumption by reusing the device buffer. Lastly, we propose accelerator-assisted gradient compression/decompression to enhance the scalability of Smart-Infinity. When scaling to multiple near-storage processing devices, the write traffic on the shared channel becomes the bottleneck. To alleviate this, we compress the gradients on the GPU and decompress them on the accelerators. It provides further acceleration from reduced traffic. As a result, Smart-Infinity achieves a significant speedup compared to the baseline. Notably, Smart-Infinity is a ready-to-use approach that is fully integrated into PyTorch on a real system. We will open-source Smart-Infinity to facilitate its use.

Jaewon Jung, Hongsun Jang, Jaeyong Song et al.

Adversarial robustness of the neural network is a significant concern when it is applied to security-critical domains. In this situation, adversarial distillation is a promising option which aims to distill the robustness of the teacher network to improve the robustness of a small student network. Previous works pretrain the teacher network to make it robust against the adversarial examples aimed at itself. However, the adversarial examples are dependent on the parameters of the target network. The fixed teacher network inevitably degrades its robustness against the unseen transferred adversarial examples which target the parameters of the student network in the adversarial distillation process. We propose PeerAiD to make a peer network learn the adversarial examples of the student network instead of adversarial examples aimed at itself. PeerAiD is an adversarial distillation that trains the peer network and the student network simultaneously in order to specialize the peer network for defending the student network. We observe that such peer networks surpass the robustness of the pretrained robust teacher model against adversarial examples aimed at the student network. With this peer network and adversarial distillation, PeerAiD achieves significantly higher robustness of the student network with AutoAttack (AA) accuracy by up to 1.66%p and improves the natural accuracy of the student network by up to 4.72%p with ResNet-18 on TinyImageNet dataset. Code is available at https://github.com/jaewonalive/PeerAiD.

Jaeyong Song, Hongsun Jang, Changmin Shin et al.

On-disk graph-based vector search (GVS) has become the dominant approach for serving large-scale vector databases at high recall, but prior systems struggle to sustain concurrent search and update throughput on high-dimensional workloads. We find the main cause of this in position seeking, a full graph traversal that every update performs to locate neighbors before linking the new vector into the graph. Position seeking is fundamentally heavier than a search query, and its cost is further amplified by two systemic limitations of current GVS systems, packed layouts that couple every edge fetch to a full vector load, and a static entrance graph whose entry points drift away from newly inserted regions as updates accumulate. We present NAVIS, an on-SSD GVS system that drives down position-seeking overhead through (i) a layout-supported selective vector read that breaks the packed-page coupling without losing its locality benefits, (ii) a dynamic lightweight entrance graph update mechanism that reuses traversal information already produced by concurrent updates, and (iii) an entrance graph-aware edgelist cache that concentrates capacity on high-reuse paths near refreshed entry points. Across multiple large-scale high-dimensional benchmarks, NAVIS enhances average insertion throughput by up to 2.74x and average concurrent search throughput by up to 1.37x while reducing average search latency by up to 25.26%.

Jaeyong Song, Seongyeon Park, Hongsun Jang et al.

Full-graph training of graph neural networks (GNNs) is widely used as it enables direct validation of algorithmic improvements by preserving complete neighborhood information. However, it typically requires multiple GPUs or servers, incurring substantial hardware and inter-device communication costs. While existing single-server methods reduce infrastructure requirements, they remain constrained by GPU and host memory capacity as graph sizes increase. To address this limitation, we introduce GriNNder, which is the first work to leverage storage devices to enable full-graph training even with limited memory. Because modern NVMe SSDs offer multi-terabyte capacities and bandwidths exceeding 10 GB/s, they provide an appealing option when memory resources are scarce. Yet, directly applying storage-based methods from other domains fails to address the unique access patterns and data dependencies in full-graph GNN training. GriNNder tackles these challenges by structured storage offloading (SSO), a framework that manages the GPU-host-storage hierarchy through coordinated cache, (re)gather, and bypass mechanisms. To realize the framework, we devise (i) a partition-wise caching strategy for host memory that exploits the observation on cross-partition dependencies, (ii) a regathering strategy for gradient computation that eliminates redundant storage operations, and (iii) a lightweight partitioning scheme that mitigates the memory requirements of existing graph partitioners. In experiments performed over various models and datasets, GriNNder achieves up to 9.78x speedup over state-of-the-art baselines and throughput comparable to distributed systems, enabling previously infeasible large-scale full-graph training even on a single GPU.

Hyunso Kim, Hyo Kang, Jaeyong Song

Recent advances in generative artificial intelligence (AI) are reshaping who enters entrepreneurship, but not who reaches the top of the quality distribution. Using data on over 160,000 product launches on Product Hunt, we find that entrepreneurial entry increased sharply following the public release of ChatGPT-3.5, driven disproportionately by solo entrepreneurs. This shift toward solo entry is particularly pronounced in categories that historically favored team-based ventures. However, much of this growth reflects low-commitment, experimental entry and does not translate into greater representation among the highest-quality outcomes. Team-based ventures are increasingly dominant in the top tiers of platform rankings. These findings suggest that generative AI lowers barriers to solo entrepreneurship while reinforcing team-based advantages.