Yuxiong He

Sheng Shen, Zhewei Yao, Chunyuan Li et al. · berkeley

The field of natural language processing (NLP) has made significant strides in recent years, particularly in the development of large-scale vision-language models (VLMs). These models aim to bridge the gap between text and visual information, enabling a more comprehensive understanding of multimedia data. However, as these models become larger and more complex, they also become more challenging to train and deploy. One approach to addressing this challenge is the use of sparsely-gated mixture-of-experts (MoE) techniques, which divide the model into smaller, specialized sub-models that can jointly solve a task. In this paper, we explore the effectiveness of MoE in scaling vision-language models, demonstrating its potential to achieve state-of-the-art performance on a range of benchmarks over dense models of equivalent computational cost. Our research offers valuable insights into stabilizing the training of MoE models, understanding the impact of MoE on model interpretability, and balancing the trade-offs between compute performance when scaling VLMs. We hope our work will inspire further research into the use of MoE for scaling large-scale vision-language models and other multimodal machine learning applications.

Zhewei Yao, Reza Yazdani Aminabadi, Minjia Zhang et al.

How to efficiently serve ever-larger trained natural language models in practice has become exceptionally challenging even for powerful cloud servers due to their prohibitive memory/computation requirements. In this work, we present an efficient and affordable post-training quantization approach to compress large Transformer-based models, termed as ZeroQuant. ZeroQuant is an end-to-end quantization and inference pipeline with three main components: (1) a fine-grained hardware-friendly quantization scheme for both weight and activations; (2) a novel affordable layer-by-layer knowledge distillation algorithm (LKD) even without the access to the original training data; (3) a highly-optimized quantization system backend support to remove the quantization/dequantization overhead. As such, we are able to show that: (1) ZeroQuant can reduce the precision for weights and activations to INT8 in a cost-free way for both BERT and GPT3-style models with minimal accuracy impact, which leads to up to 5.19x/4.16x speedup on those models compared to FP16 inference; (2) ZeroQuant plus LKD affordably quantize the weights in the fully-connected module to INT4 along with INT8 weights in the attention module and INT8 activations, resulting in 3x memory footprint reduction compared to the FP16 model; (3) ZeroQuant can be directly applied to two of the largest open-sourced language models, including GPT-J6B and GPT-NeoX20, for which our INT8 model achieves similar accuracy as the FP16 model but achieves up to 5.2x better efficiency.

Shuaiwen Leon Song, Bonnie Kruft, Minjia Zhang et al. · microsoft-research

In the upcoming decade, deep learning may revolutionize the natural sciences, enhancing our capacity to model and predict natural occurrences. This could herald a new era of scientific exploration, bringing significant advancements across sectors from drug development to renewable energy. To answer this call, we present DeepSpeed4Science initiative (deepspeed4science.ai) which aims to build unique capabilities through AI system technology innovations to help domain experts to unlock today's biggest science mysteries. By leveraging DeepSpeed's current technology pillars (training, inference and compression) as base technology enablers, DeepSpeed4Science will create a new set of AI system technologies tailored for accelerating scientific discoveries by addressing their unique complexity beyond the common technical approaches used for accelerating generic large language models (LLMs). In this paper, we showcase the early progress we made with DeepSpeed4Science in addressing two of the critical system challenges in structural biology research.

Zhewei Yao, Xiaoxia Wu, Conglong Li et al.

Most of the existing multi-modal models, hindered by their incapacity to adeptly manage interleaved image-and-text inputs in multi-image, multi-round dialogues, face substantial constraints in resource allocation for training and data accessibility, impacting their adaptability and scalability across varied interaction realms. To address this, we present the DeepSpeed-VisualChat framework, designed to optimize Large Language Models (LLMs) by incorporating multi-modal capabilities, with a focus on enhancing the proficiency of Large Vision and Language Models in handling interleaved inputs. Our framework is notable for (1) its open-source support for multi-round and multi-image dialogues, (2) introducing an innovative multi-modal causal attention mechanism, and (3) utilizing data blending techniques on existing datasets to assure seamless interactions in multi-round, multi-image conversations. Compared to existing frameworks, DeepSpeed-VisualChat shows superior scalability up to 70B parameter language model size, representing a significant advancement in multi-modal language models and setting a solid foundation for future explorations.

Reza Yazdani Aminabadi, Samyam Rajbhandari, Minjia Zhang et al.

The past several years have witnessed the success of transformer-based models, and their scale and application scenarios continue to grow aggressively. The current landscape of transformer models is increasingly diverse: the model size varies drastically with the largest being of hundred-billion parameters; the model characteristics differ due to the sparsity introduced by the Mixture-of-Experts; the target application scenarios can be latency-critical or throughput-oriented; the deployment hardware could be single- or multi-GPU systems with different types of memory and storage, etc. With such increasing diversity and the fast-evolving pace of transformer models, designing a highly performant and efficient inference system is extremely challenging. In this paper, we present DeepSpeed Inference, a comprehensive system solution for transformer model inference to address the above-mentioned challenges. DeepSpeed Inference consists of (1) a multi-GPU inference solution to minimize latency while maximizing the throughput of both dense and sparse transformer models when they fit in aggregate GPU memory, and (2) a heterogeneous inference solution that leverages CPU and NVMe memory in addition to the GPU memory and compute to enable high inference throughput with large models which do not fit in aggregate GPU memory. DeepSpeed Inference reduces latency by up to 7.3X over the state-of-the-art for latency-oriented scenarios and increases throughput by over 1.5x for throughput-oriented scenarios. Moreover, it enables trillion parameter scale inference under real-time latency constraints by leveraging hundreds of GPUs, an unprecedented scale for inference. It can inference 25x larger models than with GPU-only solutions, while delivering a high throughput of 84 TFLOPS (over $50\%$ of A6000 peak).

Zhewei Yao, Reza Yazdani Aminabadi, Olatunji Ruwase et al.

ChatGPT-like models have revolutionized various applications in artificial intelligence, from summarization and coding to translation, matching or even surpassing human performance. However, the current landscape lacks an accessible, efficient, and cost-effective end-to-end RLHF (Reinforcement Learning with Human Feedback) training pipeline for these powerful models, particularly when training at the scale of billions of parameters. This paper introduces DeepSpeed-Chat, a novel system that democratizes RLHF training, making it accessible to the AI community. DeepSpeed-Chat offers three key capabilities: an easy-to-use training and inference experience for ChatGPT-like models, a DeepSpeed-RLHF pipeline that replicates the training pipeline from InstructGPT, and a robust DeepSpeed-RLHF system that combines various optimizations for training and inference in a unified way. The system delivers unparalleled efficiency and scalability, enabling training of models with hundreds of billions of parameters in record time and at a fraction of the cost. With this development, DeepSpeed-Chat paves the way for broader access to advanced RLHF training, even for data scientists with limited resources, thereby fostering innovation and further development in the field of AI.

Sam Ade Jacobs, Masahiro Tanaka, Chengming Zhang et al.

Computation in a typical Transformer-based large language model (LLM) can be characterized by batch size, hidden dimension, number of layers, and sequence length. Until now, system works for accelerating LLM training have focused on the first three dimensions: data parallelism for batch size, tensor parallelism for hidden size and pipeline parallelism for model depth or layers. These widely studied forms of parallelism are not targeted or optimized for long sequence Transformer models. Given practical application needs for long sequence LLM, renewed attentions are being drawn to sequence parallelism. However, existing works in sequence parallelism are constrained by memory-communication inefficiency, limiting their scalability to long sequence large models. In this work, we introduce DeepSpeed-Ulysses, a novel, portable and effective methodology for enabling highly efficient and scalable LLM training with extremely long sequence length. DeepSpeed-Ulysses at its core partitions input data along the sequence dimension and employs an efficient all-to-all collective communication for attention computation. Theoretical communication analysis shows that whereas other methods incur communication overhead as sequence length increases, DeepSpeed-Ulysses maintains constant communication volume when sequence length and compute devices are increased proportionally. Furthermore, experimental evaluations show that DeepSpeed-Ulysses trains 2.5x faster with 4x longer sequence length than the existing method SOTA baseline.

Siddharth Singh, Olatunji Ruwase, Ammar Ahmad Awan et al.

Mixture-of-Experts (MoE) is a neural network architecture that adds sparsely activated expert blocks to a base model, increasing the number of parameters without impacting computational costs. However, current distributed deep learning frameworks are limited in their ability to train high-quality MoE models with large base models. In this work, we present DeepSpeed-TED, a novel, three-dimensional, hybrid parallel algorithm that combines data, tensor, and expert parallelism to enable the training of MoE models with 4 to 8x larger base models than the current state-of-the-art. We also describe memory optimizations in the optimizer step, and communication optimizations that eliminate unnecessary data movement. We implement our approach in DeepSpeed and achieve speedups of 26% over a baseline (i.e. without our communication optimizations) when training a 40 billion parameter MoE model (6.7 billion base model with 16 experts) on 128 V100 GPUs.

Lanxiang Hu, Siqi Kou, Yichao Fu et al.

Multi-token generation has emerged as a promising paradigm for accelerating transformer-based large model inference. Recent efforts primarily explore diffusion Large Language Models (dLLMs) for parallel decoding to reduce inference latency. To achieve AR-level generation quality, many techniques adapt AR models into dLLMs to enable parallel decoding. However, they suffer from limited speedup compared to AR models due to a pretrain-to-posttrain mismatch. Specifically, the masked data distribution in post-training deviates significantly from the real-world data distribution seen during pretraining, and dLLMs rely on bidirectional attention, which conflicts with the causal prior learned during pretraining and hinders the integration of exact KV cache reuse. To address this, we introduce Jacobi Forcing, a progressive distillation paradigm where models are trained on their own generated parallel decoding trajectories, smoothly shifting AR models into efficient parallel decoders while preserving their pretrained causal inference property. The models trained under this paradigm, Jacobi Forcing Model, achieves 3.8x wall-clock speedup on coding and math benchmarks with minimal loss in performance. Based on Jacobi Forcing Models' trajectory characteristics, we introduce multi-block decoding with rejection recycling, which enables up to 4.5x higher token acceptance count per iteration and nearly 4.0x wall-clock speedup, effectively trading additional compute for lower inference latency. Our code is available at https://github.com/hao-ai-lab/JacobiForcing.

Xiaoxia Wu, Cheng Li, Reza Yazdani Aminabadi et al.

Improving the deployment efficiency of transformer-based language models has been challenging given their high computation and memory cost. While INT8 quantization has recently been shown to be effective in reducing both the memory cost and latency while preserving model accuracy, it remains unclear whether we can leverage INT4 (which doubles peak hardware throughput) to achieve further latency improvement. In this study, we explore the feasibility of employing INT4 weight and activation (W4A4) quantization for language models. Our findings indicate that W4A4 quantization introduces no to negligible accuracy degradation for encoder-only and encoder-decoder models, but causes a significant accuracy drop for decoder-only models. To materialize the performance gain using W4A4, we develop a highly optimized end-to-end W4A4 encoder inference pipeline supporting different quantization strategies. Our INT4 pipeline is $8.5\times$ faster for latency-oriented scenarios and up to $3\times$ for throughput-oriented scenarios compared to the inference of FP16, and improves the SOTA BERT INT8 performance from FasterTransformer by up to $1.7\times$. We provide insights into the failure cases when applying W4A4 to decoder-only models, and further explore the compatibility of INT4 quantization with other compression methods, like pruning and layer reduction.

Guanhua Wang, Heyang Qin, Sam Ade Jacobs et al.

Zero Redundancy Optimizer (ZeRO) has been used to train a wide range of large language models on massive GPUs clusters due to its ease of use, efficiency, and good scalability. However, when training on low-bandwidth clusters, or at scale which forces batch size per GPU to be small, ZeRO's effective throughput is limited because of high communication volume from gathering weights in forward pass, backward pass, and averaging gradients. This paper introduces three communication volume reduction techniques, which we collectively refer to as ZeRO++, targeting each of the communication collectives in ZeRO. First is block-quantization based all-gather. Second is data remapping that trades-off communication for more memory. Third is a novel all-to-all based quantized gradient averaging paradigm as replacement of reduce-scatter collective, which preserves accuracy despite communicating low precision data. Collectively, ZeRO++ reduces communication volume of ZeRO by 4x, enabling up to 2.16x better throughput at 384 GPU scale.

Xiaoxia Wu, Zhewei Yao, Minjia Zhang et al.

Extreme compression, particularly ultra-low bit precision (binary/ternary) quantization, has been proposed to fit large NLP models on resource-constraint devices. However, to preserve the accuracy for such aggressive compression schemes, cutting-edge methods usually introduce complicated compression pipelines, e.g., multi-stage expensive knowledge distillation with extensive hyperparameter tuning. Also, they oftentimes focus less on smaller transformer models that have already been heavily compressed via knowledge distillation and lack a systematic study to show the effectiveness of their methods. In this paper, we perform a very comprehensive systematic study to measure the impact of many key hyperparameters and training strategies from previous works. As a result, we find out that previous baselines for ultra-low bit precision quantization are significantly under-trained. Based on our study, we propose a simple yet effective compression pipeline for extreme compression, named XTC. XTC demonstrates that (1) we can skip the pre-training knowledge distillation to obtain a 5-layer BERT while achieving better performance than previous state-of-the-art methods, e.g., the 6-layer TinyBERT; (2) extreme quantization plus layer reduction is able to reduce the model size by 50x, resulting in new state-of-the-art results on GLUE tasks.

Quentin Anthony, Ammar Ahmad Awan, Jeff Rasley et al.

In recent years, the training requirements of many state-of-the-art Deep Learning (DL) models have scaled beyond the compute and memory capabilities of a single processor, and necessitated distribution among processors. Training such massive models necessitates advanced parallelism strategies to maintain efficiency. However, such distributed DL parallelism strategies require a varied mixture of collective and point-to-point communication operations across a broad range of message sizes and scales. Examples of models using advanced parallelism strategies include Deep Learning Recommendation Models (DLRM) and Mixture-of-Experts (MoE). Communication libraries' performance varies wildly across different communication operations, scales, and message sizes. We propose MCR-DL: an extensible DL communication framework that supports all point-to-point and collective operations while enabling users to dynamically mix-and-match communication backends for a given operation without deadlocks. MCR-DL also comes packaged with a tuning suite for dynamically selecting the best communication backend for a given input tensor. We select DeepSpeed-MoE and DLRM as candidate DL models and demonstrate a 31% improvement in DS-MoE throughput on 256 V100 GPUs on the Lassen HPC system. Further, we achieve a 20% throughput improvement in a dense Megatron-DeepSpeed model and a 25% throughput improvement in DLRM on 32 A100 GPUs with the Theta-GPU HPC system.

Zhewei Yao, Xiaoxia Wu, Conglong Li et al.

Large-scale transformer models have become the de-facto architectures for various machine learning applications, e.g., CV and NLP. However, those large models also introduce prohibitive training costs. To mitigate this issue, we propose a novel random and layerwise token dropping method (random-LTD), which skips the computation of a subset of the input tokens at all middle layers. Particularly, random-LTD achieves considerable speedups and comparable accuracy as the standard training baseline. Compared to other token dropping methods, random-LTD does not require (1) any importance score-based metrics, (2) any special token treatment (e.g., [CLS]), and (3) many layers in full sequence length training except the first and the last layers. Besides, a new LayerToken learning rate schedule is proposed for pretraining problems that resolve the heavy tuning requirement for our proposed training mechanism. Finally, we demonstrate that random-LTD can be applied to broader applications, including GPT and BERT pretraining as well as ViT and GPT finetuning tasks. Our results show that random-LTD can save about 33.3% theoretical compute cost and 25.6% wall-clock training time while achieving similar zero-shot evaluations on GPT-31.3B as compared to baseline.

Yuxin Ma, Ping Gong, Jun Yi et al.

Graph Neural Networks (GNNs) is a promising approach for applications with nonEuclidean data. However, training GNNs on large scale graphs with hundreds of millions nodes is both resource and time consuming. Different from DNNs, GNNs usually have larger memory footprints, and thus the GPU memory capacity and PCIe bandwidth are the main resource bottlenecks in GNN training. To address this problem, we present BiFeat: a graph feature quantization methodology to accelerate GNN training by significantly reducing the memory footprint and PCIe bandwidth requirement so that GNNs can take full advantage of GPU computing capabilities. Our key insight is that unlike DNN, GNN is less prone to the information loss of input features caused by quantization. We identify the main accuracy impact factors in graph feature quantization and theoretically prove that BiFeat training converges to a network where the loss is within $ε$ of the optimal loss of uncompressed network. We perform extensive evaluation of BiFeat using several popular GNN models and datasets, including GraphSAGE on MAG240M, the largest public graph dataset. The results demonstrate that BiFeat achieves a compression ratio of more than 30 and improves GNN training speed by 200%-320% with marginal accuracy loss. In particular, BiFeat achieves a record by training GraphSAGE on MAG240M within one hour using only four GPUs.

Connor Holmes, Minjia Zhang, Yuxiong He et al.

In recent years, large pre-trained Transformer networks have demonstrated dramatic improvements in many natural language understanding tasks. However, the huge size of these models brings significant challenges to their fine-tuning and online deployment due to latency and cost constraints. New hardware supporting both N:M semi-structured sparsity and low-precision integer computation is a promising solution to boost DNN model serving efficiency. However, there have been very few studies that systematically investigate to what extent pre-trained Transformer networks benefit from the combination of these techniques, as well as how to best compress each component of the Transformer. We propose a flexible compression framework NxMiFormer that performs simultaneous sparsification and quantization using ADMM and STE-based QAT. Furthermore, we present and inexpensive, heuristic-driven search algorithm that identifies promising heterogeneous compression configurations that meet a compression ratio constraint. When evaluated across the GLUE suite of NLU benchmarks, our approach can achieve up to 93% compression of the encoders of a BERT model while retaining 98.2% of the original model accuracy and taking full advantage of the hardware's capabilities. Heterogeneous configurations found the by the search heuristic maintain 99.5% of the baseline accuracy while still compressing the model by 87.5%.

Chengming Zhang, Shaden Smith, Baixi Sun et al.

Collaborative filtering (CF) has been proven to be one of the most effective techniques for recommendation. Among all CF approaches, SimpleX is the state-of-the-art method that adopts a novel loss function and a proper number of negative samples. However, there is no work that optimizes SimpleX on multi-core CPUs, leading to limited performance. To this end, we perform an in-depth profiling and analysis of existing SimpleX implementations and identify their performance bottlenecks including (1) irregular memory accesses, (2) unnecessary memory copies, and (3) redundant computations. To address these issues, we propose an efficient CF training system (called HEAT) that fully enables the multi-level caching and multi-threading capabilities of modern CPUs. Specifically, the optimization of HEAT is threefold: (1) It tiles the embedding matrix to increase data locality and reduce cache misses (thus reduces read latency); (2) It optimizes stochastic gradient descent (SGD) with sampling by parallelizing vector products instead of matrix-matrix multiplications, in particular the similarity computation therein, to avoid memory copies for matrix data preparation; and (3) It aggressively reuses intermediate results from the forward phase in the backward phase to alleviate redundant computation. Evaluation on five widely used datasets with both x86- and ARM-architecture processors shows that HEAT achieves up to 45.2X speedup over existing CPU solution and 4.5X speedup and 7.9X cost reduction in Cloud over existing GPU solution with NVIDIA V100 GPU.

Zhewei Yao, Reza Yazdani Aminabadi, Stephen Youn et al.

Quantization techniques are pivotal in reducing the memory and computational demands of deep neural network inference. Existing solutions, such as ZeroQuant, offer dynamic quantization for models like BERT and GPT but overlook crucial memory-bounded operators and the complexities of per-token quantization. Addressing these gaps, we present a novel, fully hardware-enhanced robust optimized post-training W8A8 quantization framework, ZeroQuant-HERO. This framework uniquely integrates both memory bandwidth and compute-intensive operators, aiming for optimal hardware performance. Additionally, it offers flexibility by allowing specific INT8 modules to switch to FP16/BF16 mode, enhancing accuracy.

Bohan Zhai, Canwen Xu, Yuxiong He et al.

Text-to-SQL demands precise reasoning to convert natural language questions into structured queries. While large language models (LLMs) excel in many reasoning tasks, their ability to leverage Chain-of-Thought (CoT) reasoning for text-to-SQL remains underexplored. We identify critical limitations: zero-shot CoT offers minimal gains, and Direct Preference Optimization (DPO) applied without CoT yields marginal improvements. We propose ExCoT, a novel framework that iteratively optimizes open-source LLMs by combining CoT reasoning with off-policy and on-policy DPO, relying solely on execution accuracy as feedback. This approach eliminates the need for reward models or human-annotated preferences. Our experimental results demonstrate significant performance gains: ExCoT improves execution accuracy on BIRD dev set from 57.37% to 68.51% and on Spider test set from 78.81% to 86.59% for LLaMA-3 70B, with Qwen-2.5-Coder demonstrating similar improvements. Our best model achieves state-of-the-art performance in the single-model setting on both BIRD and Spider datasets, notably achieving 68.53% on the BIRD test set.

Samyam Rajbhandari, Mert Hidayetoglu, Aurick Qiao et al.

Inference is now the dominant AI workload, yet existing systems force trade-offs between latency, throughput, and cost. Arctic Inference, an open-source vLLM plugin from Snowflake AI Research, introduces Shift Parallelism, a dynamic parallelism strategy that adapts to real-world traffic while integrating speculative decoding, SwiftKV compute reduction, and optimized embedding inference. It achieves up to 3.4 times faster request completion, 1.75 times faster generation, and 1.6M tokens/sec per GPU for embeddings, outperforming both latency- and throughput-optimized deployments. Already powering Snowflake Cortex AI, Arctic Inference delivers state-of-the-art, cost-effective inference for enterprise AI and is now available to the community.

Stas Bekman, Samyam Rajbhandari, Michael Wyatt et al.

Long sequences are critical for applications like RAG, long document summarization, multi-modality, etc., and modern LLMs, like Llama 4 Scout, support max sequence length of up to 10 million tokens. However, outside of enterprise labs, long sequence training is challenging for the AI community with limited system support in the open-source space. Out-of-box, even on a modern NVIDIA H100 80GB GPU cluster, training Llama 8B model with sequence over 32K runs out of memory on a basic Hugging Face (HF) model due to two reasons: i) LLM training workloads are not optimized to fully leverage a single GPU memory, ii) existing solutions for leveraging multiple GPU memory are not easily available to HF models, making long sequence training inaccessible. We address this with Arctic Long Sequence Training (ALST). It offers a combination of attention-agnostic single GPU and multi-GPU memory optimizations, that enables it to support out-of-box training of multi-million sequence length for a wide variety of HF models. ALST supports training Meta's Llama 8B model with 500K sequence length on a single H100 GPU, 3.7M on a single 8xH100 GPU node, and over 15M on a 4 node cluster, an increase of over 400x compared to the 32K baseline for the latter. ALST is fully compatible with HF models and open-sourced via Deepspeed https://www.deepspeed.ai/tutorials/ulysses-alst-sequence-pallellism/ and Arctic Training https://github.com/snowflakedb/ArcticTraining/blob/main/projects/sequence-parallelism/README.md.

Haojun Xia, Zhen Zheng, Xiaoxia Wu et al.

Six-bit quantization (FP6) can effectively reduce the size of large language models (LLMs) and preserve the model quality consistently across varied applications. However, existing systems do not provide Tensor Core support for FP6 quantization and struggle to achieve practical performance improvements during LLM inference. It is challenging to support FP6 quantization on GPUs due to (1) unfriendly memory access of model weights with irregular bit-width and (2) high runtime overhead of weight de-quantization. To address these problems, we propose TC-FPx, the first full-stack GPU kernel design scheme with unified Tensor Core support of float-point weights for various quantization bit-width. We integrate TC-FPx kernel into an existing inference system, providing new end-to-end support (called FP6-LLM) for quantized LLM inference, where better trade-offs between inference cost and model quality are achieved. Experiments show that FP6-LLM enables the inference of LLaMA-70b using only a single GPU, achieving 1.69x-2.65x higher normalized inference throughput than the FP16 baseline. The source code is publicly available at https://github.com/usyd-fsalab/fp6_llm.

Young Jin Kim, Ammar Ahmad Awan, Alexandre Muzio et al.

The Mixture of Experts (MoE) models are an emerging class of sparsely activated deep learning models that have sublinear compute costs with respect to their parameters. In contrast with dense models, the sparse architecture of MoE offers opportunities for drastically growing model size with significant accuracy gain while consuming much lower compute budget. However, supporting large scale MoE training also has its own set of system and modeling challenges. To overcome the challenges and embrace the opportunities of MoE, we first develop a system capable of scaling MoE models efficiently to trillions of parameters. It combines multi-dimensional parallelism and heterogeneous memory technologies harmoniously with MoE to empower 8x larger models on the same hardware compared with existing work. Besides boosting system efficiency, we also present new training methods to improve MoE sample efficiency and leverage expert pruning strategy to improve inference time efficiency. By combining the efficient system and training methods, we are able to significantly scale up large multitask multilingual models for language generation which results in a great improvement in model accuracy. A model trained with 10 billion parameters on 50 languages can achieve state-of-the-art performance in Machine Translation (MT) and multilingual natural language generation tasks. The system support of efficient MoE training has been implemented and open-sourced with the DeepSpeed library.

Samyam Rajbhandari, Olatunji Ruwase, Jeff Rasley et al.

In the last three years, the largest dense deep learning models have grown over 1000x to reach hundreds of billions of parameters, while the GPU memory has only grown by 5x (16 GB to 80 GB). Therefore, the growth in model scale has been supported primarily though system innovations that allow large models to fit in the aggregate GPU memory of multiple GPUs. However, we are getting close to the GPU memory wall. It requires 800 NVIDIA V100 GPUs just to fit a trillion parameter model for training, and such clusters are simply out of reach for most data scientists. In addition, training models at that scale requires complex combinations of parallelism techniques that puts a big burden on the data scientists to refactor their model. In this paper we present ZeRO-Infinity, a novel heterogeneous system technology that leverages GPU, CPU, and NVMe memory to allow for unprecedented model scale on limited resources without requiring model code refactoring. At the same time it achieves excellent training throughput and scalability, unencumbered by the limited CPU or NVMe bandwidth. ZeRO-Infinity can fit models with tens and even hundreds of trillions of parameters for training on current generation GPU clusters. It can be used to fine-tune trillion parameter models on a single NVIDIA DGX-2 node, making large models more accessible. In terms of training throughput and scalability, it sustains over 25 petaflops on 512 NVIDIA V100 GPUs(40% of peak), while also demonstrating super linear scalability. An open source implementation of ZeRO-Infinity is available through DeepSpeed, a deep learning optimization library that makes distributed training easy, efficient, and effective.

Wei Wen, Yuxiong He, Samyam Rajbhandari et al.

Model compression is significant for the wide adoption of Recurrent Neural Networks (RNNs) in both user devices possessing limited resources and business clusters requiring quick responses to large-scale service requests. This work aims to learn structurally-sparse Long Short-Term Memory (LSTM) by reducing the sizes of basic structures within LSTM units, including input updates, gates, hidden states, cell states and outputs. Independently reducing the sizes of basic structures can result in inconsistent dimensions among them, and consequently, end up with invalid LSTM units. To overcome the problem, we propose Intrinsic Sparse Structures (ISS) in LSTMs. Removing a component of ISS will simultaneously decrease the sizes of all basic structures by one and thereby always maintain the dimension consistency. By learning ISS within LSTM units, the obtained LSTMs remain regular while having much smaller basic structures. Based on group Lasso regularization, our method achieves 10.59x speedup without losing any perplexity of a language modeling of Penn TreeBank dataset. It is also successfully evaluated through a compact model with only 2.69M weights for machine Question Answering of SQuAD dataset. Our approach is successfully extended to non- LSTM RNNs, like Recurrent Highway Networks (RHNs). Our source code is publicly available at https://github.com/wenwei202/iss-rnns

Connor Holmes, Masahiro Tanaka, Michael Wyatt et al. · microsoft-research

The deployment and scaling of large language models (LLMs) have become critical as they permeate various applications, demanding high-throughput and low-latency serving systems. Existing frameworks struggle to balance these requirements, especially for workloads with long prompts. This paper introduces DeepSpeed-FastGen, a system that employs Dynamic SplitFuse, a novel prompt and generation composition strategy, to deliver up to 2.3x higher effective throughput, 2x lower latency on average, and up to 3.7x lower (token-level) tail latency, compared to state-of-the-art systems like vLLM. We leverage a synergistic combination of DeepSpeed-MII and DeepSpeed-Inference to provide an efficient and easy-to-use serving system for LLMs. DeepSpeed-FastGen's advanced implementation supports a range of models and offers both non-persistent and persistent deployment options, catering to diverse user scenarios from interactive sessions to long-running applications. We present a detailed benchmarking methodology, analyze the performance through latency-throughput curves, and investigate scalability via load balancing. Our evaluations demonstrate substantial improvements in throughput and latency across various models and hardware configurations. We discuss our roadmap for future enhancements, including broader model support and new hardware backends. The DeepSpeed-FastGen code is readily available for community engagement and contribution.

Xiaoxia Wu, Haojun Xia, Stephen Youn et al. · microsoft-research

This study examines 4-bit quantization methods like GPTQ in large language models (LLMs), highlighting GPTQ's overfitting and limited enhancement in Zero-Shot tasks. While prior works merely focusing on zero-shot measurement, we extend task scope to more generative categories such as code generation and abstractive summarization, in which we found that INT4 quantization can significantly underperform. However, simply shifting to higher precision formats like FP6 has been particularly challenging, thus overlooked, due to poor performance caused by the lack of sophisticated integration and system acceleration strategies on current AI hardware. Our results show that FP6, even with a coarse-grain quantization scheme, performs robustly across various algorithms and tasks, demonstrating its superiority in accuracy and versatility. Notably, with the FP6 quantization, \codestar-15B model performs comparably to its FP16 counterpart in code generation, and for smaller models like the 406M it closely matches their baselines in summarization. Neither can be achieved by INT4. To better accommodate various AI hardware and achieve the best system performance, we propose a novel 4+2 design for FP6 to achieve similar latency to the state-of-the-art INT4 fine-grain quantization. With our design, FP6 can become a promising solution to the current 4-bit quantization methods used in LLMs.

Zhewei Yao, Guoheng Sun, Lukasz Borchmann et al.

Translating natural language into SQL (Test2SQL) is a longstanding challenge at the intersection of natural language understanding and structured data access. While large language models (LLMs) have significantly improved fluency in SQL generation, producing correct and executable SQL--particularly for complex queries--remains a bottleneck. We present Arctic-Text2SQL-R1, a reinforcement learning (RL) framework and model family designed to generate accurate, executable SQL using a lightweight reward signal based solely on execution correctness. Our approach avoids brittle intermediate supervision and complex reward shaping, promoting stable training and alignment with the end task. Combined with carefully curated data, strong supervised initialization, and effective training practices, Arctic-Text2SQL-R1 achieves state-of-the-art execution accuracy across six diverse Test2SQL benchmarks, including the top position on the BIRD leaderboard. Notably, our 7B model outperforms prior 70B-class systems, highlighting the framework's scalability and efficiency. We further demonstrate inference-time robustness through simple extensions like value retrieval and majority voting. Extensive experiments and ablation studies offer both positive and negative insights, providing practical guidance for future Test2SQL research.

Guanhua Wang, Olatunji Ruwase, Bing Xie et al.

Model checkpoints are critical Deep Learning (DL) artifacts that enable fault tolerance for training and downstream applications, such as inference. However, writing checkpoints to persistent storage, and other I/O aspects of DL training, are mostly ignored by compute-focused optimization efforts for faster training of rapidly growing models and datasets. Towards addressing this imbalance, we propose FastPersist to accelerate checkpoint creation in DL training. FastPersist combines three novel techniques: (i) NVMe optimizations for faster checkpoint writes to SSDs, (ii) efficient write parallelism using the available SSDs in training environments, and (iii) overlapping checkpointing with independent training computations. Our evaluation using real world dense and sparse DL models shows that FastPersist creates checkpoints in persistent storage up to 116x faster than baseline, and enables per-iteration checkpointing with negligible overhead.

Fengxiang Bie, Yibo Yang, Zhongzhu Zhou et al.

Text-to-image generation (TTI) refers to the usage of models that could process text input and generate high fidelity images based on text descriptions. Text-to-image generation using neural networks could be traced back to the emergence of Generative Adversial Network (GAN), followed by the autoregressive Transformer. Diffusion models are one prominent type of generative model used for the generation of images through the systematic introduction of noises with repeating steps. As an effect of the impressive results of diffusion models on image synthesis, it has been cemented as the major image decoder used by text-to-image models and brought text-to-image generation to the forefront of machine-learning (ML) research. In the era of large models, scaling up model size and the integration with large language models have further improved the performance of TTI models, resulting the generation result nearly indistinguishable from real-world images, revolutionizing the way we retrieval images. Our explorative study has incentivised us to think that there are further ways of scaling text-to-image models with the combination of innovative model architectures and prediction enhancement techniques. We have divided the work of this survey into five main sections wherein we detail the frameworks of major literature in order to delve into the different types of text-to-image generation methods. Following this we provide a detailed comparison and critique of these methods and offer possible pathways of improvement for future work. In the future work, we argue that TTI development could yield impressive productivity improvements for creation, particularly in the context of the AIGC era, and could be extended to more complex tasks such as video generation and 3D generation.

Pareesa Ameneh Golnari, Zhewei Yao, Yuxiong He

This study examines the impact of optimizing the Stable Diffusion (SD) guided inference pipeline. We propose optimizing certain denoising steps by limiting the noise computation to conditional noise and eliminating unconditional noise computation, thereby reducing the complexity of the target iterations by 50%. Additionally, we demonstrate that later iterations of the SD are less sensitive to optimization, making them ideal candidates for applying the suggested optimization. Our experiments show that optimizing the last 20% of the denoising loop iterations results in an 8.2% reduction in inference time with almost no perceivable changes to the human eye. Furthermore, we found that by extending the optimization to 50% of the last iterations, we can reduce inference time by approximately 20.3%, while still generating visually pleasing images.

Yucheng Lu, Conglong Li, Minjia Zhang et al.

1-bit gradient compression and local steps are two representative techniques that enable drastic communication reduction in distributed SGD. Their benefits, however, remain an open question on Adam-based large model pre-training (e.g. BERT and GPT). In this paper, we demonstrate the non-linearity in Adam causes slow convergence even when 1-bit compression or local steps are individually applied. To alleviate this limitation, we propose 0/1 Adam that linearizes each Adam step via approximating its optimizer states using their stale estimates and linear correlation. 0/1 Adam performs an Adam-like step to preserve the adaptivity, while its linearity allows utilizing 1-bit compression and local steps simultaneously for wall-clock time speed up. We provide convergence guarantee for 0/1 Adam on smooth non-convex objectives. On various large-scale benchmarks such as BERT-Base, BERT-Large, GPT-2 pre-training and ImageNet, we demonstrate on up to 128 GPUs that 0/1 Adam is able to reduce up to 87% of data volume, 54% of communication rounds, and achieve up to 2$\times$ higher training throughput and end-to-end training time reduction compared to the state-of-the-art baseline 1-bit Adam; while enjoying the same statistical convergence speed and end task model accuracy on GLUE dataset and ImageNet validation set.

Minjia Zhang, Niranjan Uma Naresh, Yuxiong He

In recent years, large pre-trained Transformer-based language models have led to dramatic improvements in many natural language understanding tasks. To train these models with increasing sizes, many neural network practitioners attempt to increase the batch sizes in order to leverage multiple GPUs to improve training speed. However, increasing the batch size often makes the optimization more difficult, leading to slow convergence or poor generalization that can require orders of magnitude more training time to achieve the same model quality. In this paper, we explore the steepness of the loss landscape of large-batch optimization for adapting pre-trained Transformer-based language models to domain-specific tasks and find that it tends to be highly complex and irregular, posing challenges to generalization on downstream tasks. To tackle this challenge, we propose ScaLA, a novel and efficient method to accelerate the adaptation speed of pre-trained transformer networks. Different from prior methods, we take a sequential game-theoretic approach by adding lightweight adversarial noise into large-batch optimization, which significantly improves adaptation speed while preserving model generalization. Experiment results show that ScaLA attains 2.7--9.8$\times$ adaptation speedups over the baseline for GLUE on BERT-base and RoBERTa-large, while achieving comparable and sometimes higher accuracy than the state-of-the-art large-batch optimization methods. Finally, we also address the theoretical aspect of large-batch optimization with adversarial noise and provide a theoretical convergence rate analysis for ScaLA using techniques for analyzing non-convex saddle-point problems.

Shaden Smith, Mostofa Patwary, Brandon Norick et al.

Pretrained general-purpose language models can achieve state-of-the-art accuracies in various natural language processing domains by adapting to downstream tasks via zero-shot, few-shot and fine-tuning techniques. Because of their success, the size of these models has increased rapidly, requiring high-performance hardware, software, and algorithmic techniques to enable training such large models. As the result of a joint effort between Microsoft and NVIDIA, we present details on the training of the largest monolithic transformer based language model, Megatron-Turing NLG 530B (MT-NLG), with 530 billion parameters. In this paper, we first focus on the infrastructure as well as the 3D parallelism methodology used to train this model using DeepSpeed and Megatron. Next, we detail the training process, the design of our training corpus, and our data curation techniques, which we believe is a key ingredient to the success of the model. Finally, we discuss various evaluation results, as well as other interesting observations and new properties exhibited by MT-NLG. We demonstrate that MT-NLG achieves superior zero-, one-, and few-shot learning accuracies on several NLP benchmarks and establishes new state-of-the-art results. We believe that our contributions will help further the development of large-scale training infrastructures, large-scale language models, and natural language generations.

Samyam Rajbhandari, Conglong Li, Zhewei Yao et al.

As the training of giant dense models hits the boundary on the availability and capability of the hardware resources today, Mixture-of-Experts (MoE) models become one of the most promising model architectures due to their significant training cost reduction compared to a quality-equivalent dense model. Its training cost saving is demonstrated from encoder-decoder models (prior works) to a 5x saving for auto-aggressive language models (this work along with parallel explorations). However, due to the much larger model size and unique architecture, how to provide fast MoE model inference remains challenging and unsolved, limiting its practical usage. To tackle this, we present DeepSpeed-MoE, an end-to-end MoE training and inference solution as part of the DeepSpeed library, including novel MoE architecture designs and model compression techniques that reduce MoE model size by up to 3.7x, and a highly optimized inference system that provides 7.3x better latency and cost compared to existing MoE inference solutions. DeepSpeed-MoE offers an unprecedented scale and efficiency to serve massive MoE models with up to 4.5x faster and 9x cheaper inference compared to quality-equivalent dense models. We hope our innovations and systems help open a promising path to new directions in the large model landscape, a shift from dense to sparse MoE models, where training and deploying higher-quality models with fewer resources becomes more widely possible.

Connor Holmes, Minjia Zhang, Yuxiong He et al.

Natural Language Processing (NLP) has recently achieved success by using huge pre-trained Transformer networks. However, these models often contain hundreds of millions or even billions of parameters, bringing challenges to online deployment due to latency constraints. Recently, hardware manufacturers have introduced dedicated hardware for NxM sparsity to provide the flexibility of unstructured pruning with the runtime efficiency of structured approaches. NxM sparsity permits arbitrarily selecting M parameters to retain from a contiguous group of N in the dense representation. However, due to the extremely high complexity of pre-trained models, the standard sparse fine-tuning techniques often fail to generalize well on downstream tasks, which have limited data resources. To address such an issue in a principled manner, we introduce a new learning framework, called NxMTransformer, to induce NxM semi-structured sparsity on pretrained language models for natural language understanding to obtain better performance. In particular, we propose to formulate the NxM sparsity as a constrained optimization problem and use Alternating Direction Method of Multipliers (ADMM) to optimize the downstream tasks while taking the underlying hardware constraints into consideration. ADMM decomposes the NxM sparsification problem into two sub-problems that can be solved sequentially, generating sparsified Transformer networks that achieve high accuracy while being able to effectively execute on newly released hardware. We apply our approach to a wide range of NLP tasks, and our proposed method is able to achieve 1.7 points higher accuracy in GLUE score than current practices. Moreover, we perform detailed analysis on our approach and shed light on how ADMM affects fine-tuning accuracy for downstream tasks. Finally, we illustrate how NxMTransformer achieves performance improvement with knowledge distillation.

Conglong Li, Minjia Zhang, Yuxiong He

Recent works have demonstrated great success in pre-training large-scale autoregressive language models on massive GPUs. To reduce the wall-clock training time, a common practice is to increase the batch size and learning rate. However, such practice is often brittle and leads to a so-called stability-efficiency dilemma: increasing the batch sizes and learning rates leads to better training efficiency but can also result in training instability, leading to poor generalization accuracy or failed runs. To better understand this phenomenon, we conduct an in-depth analysis on large-scale pre-training experiments replicating the GPT-2 model. We find that there is a strong correlation between training instability and extreme values of gradient variance, and that samples with long sequence lengths contribute to these extreme gradient variance values, especially at the beginning of the training, indicating that long sequence length can be a main source of training instability. Based on the analysis, we present a Sequence Length Warmup method that aims to solve the training stability-efficiency dilemma. Experiments replicating GPT-2 models show that our approach enables stable training with 8x larger batch size and 4x larger learning rate, whereas the baseline approach struggles with training instability. To achieve the same or better zero-shot evaluation results, our method reduces the required number of training tokens and wall clock time by up to 2.2x and 3.7x, respectively. Experiments replicating GPT-3 model (125M) show that our approach enables stable training with 8x larger batch size and 40x larger learning rate, and retains 99% of the zero-shot accuracy on 11 tasks using 10x less data and 17x less time compared to the original GPT-3 training recipe, while the baseline diverges under the same settings and only retain 95% of accuracy under lower learning rate.

Zhewei Yao, Xiaoxia Wu, Linjian Ma et al.

Pruning is an effective method to reduce the memory footprint and computational cost associated with large natural language processing models. However, current pruning algorithms either only focus on one pruning category, e.g., structured pruning and unstructured, or need extensive hyperparameter tuning in order to get reasonable accuracy performance. To address these challenges, we propose LEArnable Pruning (LEAP), an effective method to gradually prune the model based on thresholds learned by gradient descent. Different than previous learnable pruning methods, which utilize $L_0$ or $L_1$ penalty to indirectly affect the final pruning ratio, LEAP introduces a novel regularization function, that directly interacts with the preset target pruning ratio. Moreover, in order to reduce hyperparameter tuning, a novel adaptive regularization coefficient is deployed to control the regularization penalty adaptively. With the new regularization term and its associated adaptive regularization coefficient, LEAP is able to be applied for different pruning granularity, including unstructured pruning, structured pruning, and hybrid pruning, with minimal hyperparameter tuning. We apply LEAP for BERT models on QQP/MNLI/SQuAD for different pruning settings. Our result shows that for all datasets, pruning granularity, and pruning ratios, LEAP achieves on-par or better results as compared to previous heavily hand-tuned methods.

Conglong Li, Ammar Ahmad Awan, Hanlin Tang et al.

To train large models (like BERT and GPT-3) on hundreds of GPUs, communication has become a major bottleneck, especially on commodity systems with limited-bandwidth TCP network. On one side large batch-size optimization such as LAMB algorithm was proposed to reduce the frequency of communication. On the other side, communication compression algorithms such as 1-bit Adam help to reduce the volume of each communication. However, we find that simply using one of the techniques is not sufficient to solve the communication challenge, especially under low network bandwidth. Motivated by this we aim to combine the power of large-batch optimization and communication compression, but we find that existing compression strategies cannot be directly applied to LAMB due to its unique adaptive layerwise learning rates. To this end, we design a new communication-efficient algorithm, 1-bit LAMB, which introduces a novel way to support adaptive layerwise learning rates under compression. In addition, we introduce a new system implementation for compressed communication using the NCCL backend of PyTorch distributed, which improves both usability and performance. For BERT-Large pre-training task with batch sizes from 8K to 64K, our evaluations on up to 256 GPUs demonstrate that 1-bit LAMB with NCCL-based backend is able to achieve up to 4.6x communication volume reduction, up to 2.8x end-to-end time-wise speedup, and the same sample-wise convergence speed (and same fine-tuning task accuracy) compared to uncompressed LAMB.

Hanlin Tang, Shaoduo Gan, Ammar Ahmad Awan et al.

Scalable training of large models (like BERT and GPT-3) requires careful optimization rooted in model design, architecture, and system capabilities. From a system standpoint, communication has become a major bottleneck, especially on commodity systems with standard TCP interconnects that offer limited network bandwidth. Communication compression is an important technique to reduce training time on such systems. One of the most effective methods is error-compensated compression, which offers robust convergence speed even under 1-bit compression. However, state-of-the-art error compensation techniques only work with basic optimizers like SGD and momentum SGD, which are linearly dependent on the gradients. They do not work with non-linear gradient-based optimizers like Adam, which offer state-of-the-art convergence efficiency and accuracy for models like BERT. In this paper, we propose 1-bit Adam that reduces the communication volume by up to $5\times$, offers much better scalability, and provides the same convergence speed as uncompressed Adam. Our key finding is that Adam's variance (non-linear term) becomes stable (after a warmup phase) and can be used as a fixed precondition for the rest of the training (compression phase). Experiments on up to 256 GPUs show that 1-bit Adam enables up to $3.3\times$ higher throughput for BERT-Large pre-training and up to $2.9\times$ higher throughput for SQuAD fine-tuning. In addition, we provide theoretical analysis for our proposed work.

Jie Ren, Samyam Rajbhandari, Reza Yazdani Aminabadi et al.

Large-scale model training has been a playing ground for a limited few requiring complex model refactoring and access to prohibitively expensive GPU clusters. ZeRO-Offload changes the large model training landscape by making large model training accessible to nearly everyone. It can train models with over 13 billion parameters on a single GPU, a 10x increase in size compared to popular framework such as PyTorch, and it does so without requiring any model change from the data scientists or sacrificing computational efficiency. ZeRO-Offload enables large model training by offloading data and compute to CPU. To preserve compute efficiency, it is designed to minimize the data movement to/from GPU, and reduce CPU compute time while maximizing memory savings on GPU. As a result, ZeRO-Offload can achieve 40 TFlops/GPU on a single NVIDIA V100 GPU for 10B parameter model compared to 30TF using PyTorch alone for a 1.4B parameter model, the largest that can be trained without running out of memory. ZeRO-Offload is also designed to scale on multiple-GPUs when available, offering near linear speedup on up to 128 GPUs. Additionally, it can work together with model parallelism to train models with over 70 billion parameters on a single DGX-2 box, a 4.5x increase in model size compared to using model parallelism alone. By combining compute and memory efficiency with ease-of-use, ZeRO-Offload democratizes large-scale model training making it accessible to even data scientists with access to just a single GPU.

Minjia Zhang, Yuxiong He

Recently, Transformer-based language models have demonstrated remarkable performance across many NLP domains. However, the unsupervised pre-training step of these models suffers from unbearable overall computational expenses. Current methods for accelerating the pre-training either rely on massive parallelism with advanced hardware or are not applicable to language modeling. In this work, we propose a method based on progressive layer dropping that speeds the training of Transformer-based language models, not at the cost of excessive hardware resources but from model architecture change and training technique boosted efficiency. Extensive experiments on BERT show that the proposed method achieves a 24% time reduction on average per sample and allows the pre-training to be 2.5 times faster than the baseline to get a similar accuracy on downstream tasks. While being faster, our pre-trained models are equipped with strong knowledge transferability, achieving comparable and sometimes higher GLUE score than the baseline when pre-trained with the same number of samples.

Hanlin Tang, Shaoduo Gan, Samyam Rajbhandari et al.

Adam is the important optimization algorithm to guarantee efficiency and accuracy for training many important tasks such as BERT and ImageNet. However, Adam is generally not compatible with information (gradient) compression technology. Therefore, the communication usually becomes the bottleneck for parallelizing Adam. In this paper, we propose a communication efficient {\bf A}DAM {\bf p}reconditioned {\bf M}omentum SGD algorithm-- named APMSqueeze-- through an error compensated method compressing gradients. The proposed algorithm achieves a similar convergence efficiency to Adam in term of epochs, but significantly reduces the running time per epoch. In terms of end-to-end performance (including the full-precision pre-condition step), APMSqueeze is able to provide {sometimes by up to $2-10\times$ speed-up depending on network bandwidth.} We also conduct theoretical analysis on the convergence and efficiency.

Reza Yazdani, Olatunji Ruwase, Minjia Zhang et al.

The effectiveness of Recurrent Neural Networks (RNNs) for tasks such as Automatic Speech Recognition has fostered interest in RNN inference acceleration. Due to the recurrent nature and data dependencies of RNN computations, prior work has designed customized architectures specifically tailored to the computation pattern of RNN, getting high computation efficiency for certain chosen model sizes. However, given that the dimensionality of RNNs varies a lot for different tasks, it is crucial to generalize this efficiency to diverse configurations. In this work, we identify adaptiveness as a key feature that is missing from today's RNN accelerators. In particular, we first show the problem of low resource-utilization and low adaptiveness for the state-of-the-art RNN implementations on GPU, FPGA and ASIC architectures. To solve these issues, we propose an intelligent tiled-based dispatching mechanism for increasing the adaptiveness of RNN computation, in order to efficiently handle the data dependencies. To do so, we propose Sharp as a hardware accelerator, which pipelines RNN computation using an effective scheduling scheme to hide most of the dependent serialization. Furthermore, Sharp employs dynamic reconfigurable architecture to adapt to the model's characteristics. Sharp achieves 2x, 2.8x, and 82x speedups on average, considering different RNN models and resource budgets, compared to the state-of-the-art ASIC, FPGA, and GPU implementations, respectively. Furthermore, we provide significant energy-reduction with respect to the previous solutions, due to the low power dissipation of Sharp (321 GFLOPS/Watt).

Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase et al.

Large deep learning models offer significant accuracy gains, but training billions to trillions of parameters is challenging. Existing solutions such as data and model parallelisms exhibit fundamental limitations to fit these models into limited device memory, while obtaining computation, communication and development efficiency. We develop a novel solution, Zero Redundancy Optimizer (ZeRO), to optimize memory, vastly improving training speed while increasing the model size that can be efficiently trained. ZeRO eliminates memory redundancies in data- and model-parallel training while retaining low communication volume and high computational granularity, allowing us to scale the model size proportional to the number of devices with sustained high efficiency. Our analysis on memory requirements and communication volume demonstrates: ZeRO has the potential to scale beyond 1 Trillion parameters using today's hardware. We implement and evaluate ZeRO: it trains large models of over 100B parameter with super-linear speedup on 400 GPUs, achieving throughput of 15 Petaflops. This represents an 8x increase in model size and 10x increase in achievable performance over state-of-the-art. In terms of usability, ZeRO can train large models of up to 13B parameters (e.g., larger than Megatron GPT 8.3B and T5 11B) without requiring model parallelism which is harder for scientists to apply. Last but not the least, researchers have used the system breakthroughs of ZeRO to create the world's largest language model (Turing-NLG, 17B parameters) with record breaking accuracy.

Samyam Rajbhandari, Harsh Shrivastava, Yuxiong He

Wide adoption of complex RNN based models is hindered by their inference performance, cost and memory requirements. To address this issue, we develop AntMan, combining structured sparsity with low-rank decomposition synergistically, to reduce model computation, size and execution time of RNNs while attaining desired accuracy. AntMan extends knowledge distillation based training to learn the compressed models efficiently. Our evaluation shows that AntMan offers up to 100x computation reduction with less than 1pt accuracy drop for language and machine reading comprehension models. Our evaluation also shows that for a given accuracy target, AntMan produces 5x smaller models than the state-of-art. Lastly, we show that AntMan offers super-linear speed gains compared to theoretical speedup, demonstrating its practical value on commodity hardware.

Minjia Zhang, Yuxiong He

With the advancement of machine learning and deep learning, vector search becomes instrumental to many information retrieval systems, to search and find best matches to user queries based on their semantic similarities.These online services require the search architecture to be both effective with high accuracy and efficient with low latency and memory footprint, which existing work fails to offer. We develop, Zoom, a new vector search solution that collaboratively optimizes accuracy, latency and memory based on a multiview approach. (1) A "preview" step generates a small set of good candidates, leveraging compressed vectors in memory for reduced footprint and fast lookup. (2) A "fullview" step on SSDs reranks those candidates with their full-length vector, striking high accuracy. Our evaluation shows that, Zoom achieves an order of magnitude improvements on efficiency while attaining equal or higher accuracy, comparing with the state-of-the-art.

Minjia Zhang, Xiaodong Liu, Wenhan Wang et al.

Neural language models (NLMs) have recently gained a renewed interest by achieving state-of-the-art performance across many natural language processing (NLP) tasks. However, NLMs are very computationally demanding largely due to the computational cost of the softmax layer over a large vocabulary. We observe that, in decoding of many NLP tasks, only the probabilities of the top-K hypotheses need to be calculated preciously and K is often much smaller than the vocabulary size. This paper proposes a novel softmax layer approximation algorithm, called Fast Graph Decoder (FGD), which quickly identifies, for a given context, a set of K words that are most likely to occur according to a NLM. We demonstrate that FGD reduces the decoding time by an order of magnitude while attaining close to the full softmax baseline accuracy on neural machine translation and language modeling tasks. We also prove the theoretical guarantee on the softmax approximation quality.