MARCA: Mamba Accelerator with ReConfigurable Architecture

• URL: http://arxiv.org/abs/2409.11440v1
• Date: Mon, 16 Sep 2024 15:18:33 GMT
• Title: MARCA: Mamba Accelerator with ReConfigurable Architecture
• Authors: Jinhao Li, Shan Huang, Jiaming Xu, Jun Liu, Li Ding, Ningyi Xu, Guohao Dai,
• Abstract summary: We propose a Mamba accelerator with reconfigurable architecture, MARCA. Reduction alternative PE array architecture for both linear and element-wise operations. Reusable nonlinear function unit based on the reconfigurable PE. Intra-operation and inter-operation buffer management strategy.
• Score: 16.48279181435065
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose a Mamba accelerator with reconfigurable architecture, MARCA.We propose three novel approaches in this paper. (1) Reduction alternative PE array architecture for both linear and element-wise operations. For linear operations, the reduction tree connected to PE arrays is enabled and executes the reduction operation. For element-wise operations, the reduction tree is disabled and the output bypasses. (2) Reusable nonlinear function unit based on the reconfigurable PE. We decompose the exponential function into element-wise operations and a shift operation by a fast biased exponential algorithm, and the activation function (SiLU) into a range detection and element-wise operations by a piecewise approximation algorithm. Thus, the reconfigurable PEs are reused to execute nonlinear functions with negligible accuracy loss.(3) Intra-operation and inter-operation buffer management strategy. We propose intra-operation buffer management strategy to maximize input data sharing for linear operations within operations, and inter-operation strategy for element-wise operations between operations. We conduct extensive experiments on Mamba model families with different sizes.MARCA achieves up to 463.22$\times$/11.66$\times$ speedup and up to 9761.42$\times$/242.52$\times$ energy efficiency compared to Intel Xeon 8358P CPU and NVIDIA Tesla A100 GPU implementations, respectively.

Related papers

• R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference [77.47238561728459]
  R-Sparse is a training-free activation sparsity approach capable of achieving high sparsity levels in advanced LLMs. Experiments on Llama-2/3 and Mistral models across ten diverse tasks demonstrate that R-Sparse achieves comparable performance at 50% model-level sparsity.
  arXiv  Detail & Related papers  (2025-04-28T03:30:32Z)
• CORDIC Is All You Need [0.18184027690235535]
  We present pipelined architecture with CORDIC block for linear MAC computations and nonlinear iterative Activation Functions. This approach focuses on a Reconfigurable Processing Engine (RPE) based systolic array. FPGA implementation achieves a reduction of up to 2.5 $times$ resource savings and 3 $times$ power compared to prior works.
  arXiv  Detail & Related papers  (2025-03-04T12:23:27Z)
• Joint Transmit and Pinching Beamforming for Pinching Antenna Systems (PASS): Optimization-Based or Learning-Based? [89.05848771674773]
  A novel antenna system ()-enabled downlink multi-user multiple-input single-output (MISO) framework is proposed. It consists of multiple waveguides, which equip numerous low-cost antennas, named (PAs) The positions of PAs can be reconfigured to both spanning large-scale path and space.
  arXiv  Detail & Related papers  (2025-02-12T18:54:10Z)
• CARE Transformer: Mobile-Friendly Linear Visual Transformer via Decoupled Dual Interaction [77.8576094863446]
  We propose a new detextbfCoupled dutextbfAl-interactive lineatextbfR atttextbfEntion (CARE) mechanism. We first propose an asymmetrical feature decoupling strategy that asymmetrically decouples the learning process for local inductive bias and long-range dependencies. By adopting a decoupled learning way and fully exploiting complementarity across features, our method can achieve both high efficiency and accuracy.
  arXiv  Detail & Related papers  (2024-11-25T07:56:13Z)
• EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference [49.94169109038806]
  This paper introduces EPS-MoE, a novel expert pipeline scheduler for MoE. Our results demonstrate an average 21% improvement in prefill throughput over existing parallel inference methods.
  arXiv  Detail & Related papers  (2024-10-16T05:17:49Z)
• A High-Speed Hardware Algorithm for Modulus Operation and its Application in Prime Number Calculation [0.0]
  The proposed algorithm use only addition, subtraction, logical, and bit shift operations. It addresses scalability challenges in cryptographic applications. The application of this algorithm in prime number calculation up to 500,000 shows its practical utility and performance advantages.
  arXiv  Detail & Related papers  (2024-07-17T13:24:52Z)
• Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient [57.9629676017527]
  We propose an optimization-based structural pruning on Large-Language Models. We learn the pruning masks in a probabilistic space directly by optimizing the loss of the pruned model. Our method operates for 2.7 hours with around 35GB memory for the 13B models on a single A100 GPU.
  arXiv  Detail & Related papers  (2024-06-15T09:31:03Z)
• Multi-Grid Tensorized Fourier Neural Operator for High-Resolution PDEs [93.82811501035569]
  We introduce a new data efficient and highly parallelizable operator learning approach with reduced memory requirement and better generalization. MG-TFNO scales to large resolutions by leveraging local and global structures of full-scale, real-world phenomena. We demonstrate superior performance on the turbulent Navier-Stokes equations where we achieve less than half the error with over 150x compression.
  arXiv  Detail & Related papers  (2023-09-29T20:18:52Z)
• Low-Latency Online Multiplier with Reduced Activities and Minimized Interconnect for Inner Product Arrays [0.8078491757252693]
  This paper proposes a low latency multiplier based on online or left-to-right arithmetic. Online arithmetic enables overlapping successive operations regardless of data dependency. Serial nature of the online algorithm and gradual increment/decrement of active slices minimize the interconnects and signal activities.
  arXiv  Detail & Related papers  (2023-04-06T01:22:27Z)
• UNETR++: Delving into Efficient and Accurate 3D Medical Image Segmentation [93.88170217725805]
  We propose a 3D medical image segmentation approach, named UNETR++, that offers both high-quality segmentation masks as well as efficiency in terms of parameters, compute cost, and inference speed. The core of our design is the introduction of a novel efficient paired attention (EPA) block that efficiently learns spatial and channel-wise discriminative features. Our evaluations on five benchmarks, Synapse, BTCV, ACDC, BRaTs, and Decathlon-Lung, reveal the effectiveness of our contributions in terms of both efficiency and accuracy.
  arXiv  Detail & Related papers  (2022-12-08T18:59:57Z)
• Neural-PIM: Efficient Processing-In-Memory with Neural Approximation of Peripherals [11.31429464715989]
  This paper presents a new PIM architecture to efficiently accelerate deep learning tasks. It is proposed to minimize the required A/D conversions with analog accumulation and neural approximated peripheral circuits. Evaluations on different benchmarks demonstrate that Neural-PIM can improve energy efficiency by 5.36x (1.73x) and speed up throughput by 3.43x (1.59x) without losing accuracy.
  arXiv  Detail & Related papers  (2022-01-30T16:14:49Z)
• PERMDNN: Efficient Compressed DNN Architecture with Permuted Diagonal Matrices [35.90103072918056]
  Deep neural network (DNN) has emerged as the most important and popular artificial intelligent (AI) technique. The growth of model size poses a key energy efficiency challenge for the underlying computing platform. This paper proposes PermDNN, a novel approach to generate and execute hardware-friendly structured sparse DNN models.
  arXiv  Detail & Related papers  (2020-04-23T02:26:40Z)
• ESSOP: Efficient and Scalable Stochastic Outer Product Architecture for Deep Learning [1.2019888796331233]
  Matrix-vector multiplications (MVM) and vector-vector outer product (VVOP) are the two most expensive operations associated with the training of deep neural networks (DNNs) We introduce efficient techniques to SC for weight update in DNNs with the activation functions required by many state-of-the-art networks. Our architecture reduces the computational cost by re-using random numbers and replacing certain FP multiplication operations by bit shift scaling. Hardware design of ESSOP at 14nm technology node shows that, compared to a highly pipelined FP16 multiplier, ESSOP is 82.2% and 93.7% better in energy
  arXiv  Detail & Related papers  (2020-03-25T07:54:42Z)
• Accelerating Feedforward Computation via Parallel Nonlinear Equation Solving [106.63673243937492]
  Feedforward computation, such as evaluating a neural network or sampling from an autoregressive model, is ubiquitous in machine learning. We frame the task of feedforward computation as solving a system of nonlinear equations. We then propose to find the solution using a Jacobi or Gauss-Seidel fixed-point method, as well as hybrid methods of both. Our method is guaranteed to give exactly the same values as the original feedforward computation with a reduced (or equal) number of parallelizable iterations, and hence reduced time given sufficient parallel computing power.
  arXiv  Detail & Related papers  (2020-02-10T10:11:31Z)

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.