PIM-Opt: Demystifying Distributed Optimization Algorithms on a Real-World Processing-In-Memory System

• URL: http://arxiv.org/abs/2404.07164v2
• Date: Fri, 27 Sep 2024 14:32:19 GMT
• Title: PIM-Opt: Demystifying Distributed Optimization Algorithms on a Real-World Processing-In-Memory System
• Authors: Steve Rhyner, Haocong Luo, Juan Gómez-Luna, Mohammad Sadrosadati, Jiawei Jiang, Ataberk Olgun, Harshita Gupta, Ce Zhang, Onur Mutlu,
• Abstract summary: Modern Machine Learning (ML) training on large-scale datasets is a time-consuming workload. It relies on the optimization algorithm Gradient Descent (SGD) due to its effectiveness, simplicity, and generalization performance. processor-centric architectures suffer from low performance and high energy consumption while executing ML training workloads. Processing-In-Memory (PIM) is a promising solution to alleviate the data movement bottleneck.
• Score: 21.09681871279162
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern Machine Learning (ML) training on large-scale datasets is a very time-consuming workload. It relies on the optimization algorithm Stochastic Gradient Descent (SGD) due to its effectiveness, simplicity, and generalization performance. Processor-centric architectures (e.g., CPUs, GPUs) commonly used for modern ML training workloads based on SGD are bottlenecked by data movement between the processor and memory units due to the poor data locality in accessing large datasets. As a result, processor-centric architectures suffer from low performance and high energy consumption while executing ML training workloads. Processing-In-Memory (PIM) is a promising solution to alleviate the data movement bottleneck by placing the computation mechanisms inside or near memory. Our goal is to understand the capabilities of popular distributed SGD algorithms on real-world PIM systems to accelerate data-intensive ML training workloads. To this end, we 1) implement several representative centralized parallel SGD algorithms on the real-world UPMEM PIM system, 2) rigorously evaluate these algorithms for ML training on large-scale datasets in terms of performance, accuracy, and scalability, 3) compare to conventional CPU and GPU baselines, and 4) discuss implications for future PIM hardware and highlight the need for a shift to an algorithm-hardware codesign. Our results demonstrate three major findings: 1) The UPMEM PIM system can be a viable alternative to state-of-the-art CPUs and GPUs for many memory-bound ML training workloads, especially when operations and datatypes are natively supported by PIM hardware, 2) it is important to carefully choose the optimization algorithms that best fit PIM, and 3) the UPMEM PIM system does not scale approximately linearly with the number of nodes for many data-intensive ML training workloads. We open source all our code to facilitate future research.

Related papers

• DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution [114.61347672265076]
  Development of MLLMs for real-world robots is challenging due to the typically limited computation and memory capacities available on robotic platforms. We propose a Dynamic Early-Exit Framework for Robotic Vision-Language-Action Model (DeeR) that automatically adjusts the size of the activated MLLM. DeeR demonstrates significant reductions in computational costs of LLM by 5.2-6.5x and GPU memory of LLM by 2-6x without compromising performance.
  arXiv  Detail & Related papers  (2024-11-04T18:26:08Z)
• Efficient Arbitrary Precision Acceleration for Large Language Models on GPU Tensor Cores [3.6385567224218556]
  Large language models (LLMs) have been widely applied but face challenges in efficient inference. We introduce a novel bipolar-INT data format that facilitates parallel computing and supports symmetric quantization. We implement an arbitrary precision matrix multiplication scheme that decomposes and recovers at the bit level, enabling flexible precision.
  arXiv  Detail & Related papers  (2024-09-26T14:17:58Z)
• Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment [56.44025052765861]
  Large language models (LLMs) have revolutionized Natural Language Processing (NLP), but their size creates computational bottlenecks. We introduce a novel approach to create accurate, sparse foundational versions of performant LLMs. We show a total speedup on CPUs for sparse-quantized LLaMA models of up to 8.6x.
  arXiv  Detail & Related papers  (2024-05-06T16:03:32Z)
• Federated Fine-Tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly [62.473245910234304]
  This paper takes a hardware-centric approach to explore how Large Language Models can be brought to modern edge computing systems. We provide a micro-level hardware benchmark, compare the model FLOP utilization to a state-of-the-art data center GPU, and study the network utilization in realistic conditions.
  arXiv  Detail & Related papers  (2023-10-04T20:27:20Z)
• FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUs [57.12856172329322]
  We envision a decentralized system unlocking the potential vast untapped consumer-level GPU. This system faces critical challenges, including limited CPU and GPU memory, low network bandwidth, the variability of peer and device heterogeneity.
  arXiv  Detail & Related papers  (2023-09-03T13:27:56Z)
• An Experimental Evaluation of Machine Learning Training on a Real Processing-in-Memory System [9.429605859159023]
  Training machine learning (ML) algorithms is a computationally intensive process, which is frequently memory-bound. Memory-centric computing systems, with processing-in-memory capabilities, can alleviate this data movement bottleneck. We implement several representative classic ML algorithms on a real-world general-purpose PIM architecture.
  arXiv  Detail & Related papers  (2022-07-16T09:39:53Z)
• NumS: Scalable Array Programming for the Cloud [82.827921577004]
  We present NumS, an array programming library which optimize NumPy-like expressions on task-based distributed systems. This is achieved through a novel scheduler called Load Simulated Hierarchical Scheduling (LSHS) We show that LSHS enhances performance on Ray by decreasing network load by a factor of 2x, requiring 4x less memory, and reducing execution time by 10x on the logistic regression problem.
  arXiv  Detail & Related papers  (2022-06-28T20:13:40Z)
• Machine Learning Training on a Real Processing-in-Memory System [9.286176889576996]
  Training machine learning algorithms is a computationally intensive process, which is frequently memory-bound. Memory-centric computing systems with processing-in-memory capabilities can alleviate this data movement bottleneck. Our work is the first one to evaluate training of machine learning algorithms on a real-world general-purpose PIM architecture.
  arXiv  Detail & Related papers  (2022-06-13T10:20:23Z)
• Heterogeneous Data-Centric Architectures for Modern Data-Intensive Applications: Case Studies in Machine Learning and Databases [9.927754948343326]
  processing-in-memory (PIM) is a promising execution paradigm that alleviates the data movement bottleneck in modern applications. In this paper, we show how to take advantage of the PIM paradigm for two modern data-intensive applications.
  arXiv  Detail & Related papers  (2022-05-29T13:43:17Z)
• MLPerfTM HPC: A Holistic Benchmark Suite for Scientific Machine Learning on HPC Systems [32.621917787044396]
  We introduceerf HPC, a benchmark suite of scientific machine learning training applications driven by the MLCommonsTM Association. We develop a systematic framework for their joint analysis and compare them in terms of data staging, algorithmic convergence, and compute performance. We conclude by characterizing each benchmark with respect to low-level memory, I/O, and network behavior.
  arXiv  Detail & Related papers  (2021-10-21T20:30:12Z)
• A Survey on Large-scale Machine Learning [67.6997613600942]
  Machine learning can provide deep insights into data, allowing machines to make high-quality predictions. Most sophisticated machine learning approaches suffer from huge time costs when operating on large-scale data. Large-scale Machine Learning aims to learn patterns from big data with comparable performance efficiently.
  arXiv  Detail & Related papers  (2020-08-10T06:07:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.