Many modern machine learning (ML) methods rely on embedding models to learn vector representations (embeddings) for a set of entities (embedding tables). As increasingly diverse ML applications utilize embedding models and embedding tables continue to grow in size and number, there has been a surge in the ad-hoc development of specialized frameworks targeted to train large embedding models for specific tasks. Although the scalability issues that arise in different embedding model training tasks are similar, each of these frameworks independently reinvents and customizes storage components for specific tasks, leading to substantial duplicated engineering efforts in both development and deployment. This paper presents MLKV, an efficient, extensible, and reusable data storage framework designed to address the scalability challenges in embedding model training, specifically data stall and staleness. MLKV augments disk-based key-value storage by democratizing optimizations that were previously exclusive to individual specialized frameworks and provides easy-to-use interfaces for embedding model training tasks. Extensive experiments on open-source workloads, as well as applications in eBay's payment transaction risk detection and seller payment risk detection, show that MLKV outperforms offloading strategies built on top of industrial-strength key-value stores by 1.6-12.6 ×. MLKV is open-source at https://github.com/llm-db/MLKV.

To handle graphs in which features or connections are evolving over time, a series of temporal graph neural networks (TGNNs) have been proposed. Despite the success of these TGNNs, the previous TGNN evaluations reveal several limitations regarding four critical issues: 1) inconsistent datasets, 2) inconsistent evaluation pipelines, 3) lacking workload diversity, and 4) lacking efficient comparison. Overall, there lacks an empirical study that puts TGNN models onto the same ground and compares them comprehensively. To this end, we propose Benchtemp, a general benchmark for evaluating TGNN models on various workloads. Benchtemp provides a set of benchmark datasets so that different TGNN models can be fairly compared. Further, Benchtemp engineers a standard pipeline that unifies the TGNN evaluation. With Benchtemp, we extensively compare the representative TGNN models on different tasks (e.g., link prediction and node classification) and settings (transductive and inductive), w.r.t. both effectiveness and efficiency metrics. We have made Benchtemp publicly available at https://github.com/qianghuangwhu/benchtemp and datasets at https://zenodo.org/record/8267846.

Recent years have witnessed an exponential growth of model scale in deep learning-based recommender systems---from Google's 2016 model with 1 billion parameters to the latest Facebook's model with 12 trillion parameters. Significant quality boost has come with each jump of the model capacity, which makes us believe the era of 100 trillion parameters is around the corner. However, the training of such models is challenging even within industrial scale data centers. We resolve this challenge by careful co-design of both optimization algorithm and distributed system architecture. Specifically, to ensure both the training efficiency and the training accuracy, we design a novel hybrid training algorithm, where the embedding layer and the dense neural network are handled by different synchronization mechanisms; then we build a system called Persia (short for parallel recommendation training system with hybrid acceleration) to support this hybrid training algorithm. Both theoretical demonstrations and empirical studies with up to 100 trillion parameters have been conducted to justify the system design and implementation of Persia. We make Persia publicly available (at github.com/PersiaML/Persia) so that anyone can easily train a recommender model at the scale of 100 trillion parameters.

Communication compression is a crucial technique for modern distributed learning systems to alleviate their communication bottlenecks over slower networks. Despite recent intensive studies of gradient compression for data parallel-style training, compressing the activations for models trained with pipeline parallelism is still an open problem. In this paper, we propose AQ-SGD, a novel activation compression algorithm for communication-efficient pipeline parallelism training over slow networks. Different from previous efforts in activation compression, instead of compressing activation values directly, AQ-SGD compresses the changes of the activations. This allows us to show, to the best of our knowledge for the first time, that one can still achieve $O(1/\sqrt{T})$ convergence rate for non-convex objectives under activation compression, without making assumptions on gradient unbiasedness that do not hold for deep learning models with non-linear activation functions. We then show that AQ-SGD can be optimized and implemented efficiently, without additional end-to-end runtime overhead. We evaluated AQ-SGD to fine-tune language models with up to 1.5 billion parameters, compressing activation to 2-4 bits. AQ-SGD provides up to $4.3\times$ end-to-end speed-up in slower networks, without sacrificing model quality. Moreover, we also show that AQ-SGD can be combined with state-of-the-art gradient compression algorithms to enable end-to-end communication compression: All communications between machines, including model gradients, forward activations, and backward gradients are compressed into lower precision. This provides up to $4.9\times$ end-to-end speed-up, without sacrificing model quality.

Training foundation models, such as GPT-3 and PaLM, can be extremely expensive, often involving tens of thousands of GPUs running continuously for months. These models are typically trained in specialized clusters featuring fast, homogeneous interconnects and using carefully designed software systems that support both data parallelism and model/pipeline parallelism. Such dedicated clusters can be costly and difficult to obtain. Can we instead leverage the much greater amount of decentralized, heterogeneous, and lower-bandwidth interconnected compute? Previous works examining the heterogeneous, decentralized setting focus on relatively small models that can be trained in a purely data parallel manner. State-of-the-art schemes for model parallel foundation model training, such as Megatron and Deepspeed, only consider the homogeneous data center setting. In this paper, we present the first study of training large foundation models with model parallelism in a decentralized regime over a heterogeneous network. Our key technical contribution is a scheduling algorithm that allocates different computational “tasklets” in the training of foundation models to a group of decentralized GPU devices connected by a slow heterogeneous network. We provide a formal cost model and further propose an efficient evolutionary algorithm to find the optimal allocation strategy. We conduct extensive experiments that represent different scenarios for learning over geo-distributed devices simulated using real-world network measurements. In the most extreme case, across 8 different cities spanning 3 continents, our approach is 4.8× faster than prior state-of-the-art training systems.