More Like this

1. MAESTRO: A Data-Centric Approach to Understand Reuse, Performance, and Hardware Cost of DNN Mappings

   Kwon, Hyoukjun; Chatarasi, Prasanth; Pellauer, Michael; Parashar, Angshuman; Sarkar, Vivek; Krishna, Tushar (July 2020, IEEE micro)

   The efficiency of an accelerator depends on three factors—mapping, deep neural network (DNN) layers, and hardware—constructing extremely complicated design space of DNN accelerators. To demystify such complicated design space and guide the DNN accelerator design for better efficiency, we propose an analytical cost model, MAESTRO. MAESTRO receives DNN model description and hardware resources information as a list, and mapping described in a data-centric representation we propose as inputs. The data centric representation consists of three directives that enable concise description of mappings in a compiler-friendly form. MAESTRO analyzes various forms of data reuse in an accelerator based on inputs quickly and generates more than 20 statistics including total latency, energy, throughput, etc., as outputs. MAESTRO’s fast analysis enables various optimization tools for DNN accelerators such as hardware design exploration tool we present as an example. 

   more » « less
2. A Semi-Decoupled Approach to Fast and Optimal Hardware-Software Co-Design of Neural Accelerators

   Lu, Bingqian; Yan, Zeyu; Shi, Yiyu; Ren, Shaolei (March 2022, TinyML Research Symposium 2022)

   In view of the performance limitations of fully-decoupled designs for neural architectures and accelerators, hardware-software co-design has been emerging to fully reap the benefits of flexible design spaces and optimize neural network performance. Nonetheless, such co-design also enlarges the total search space to practically infinity and presents substantial challenges. While the prior studies have been focusing on improving the search efficiency (e.g., via reinforcement learning), they commonly rely on co-searches over the entire architecture-accelerator design space. In this paper, we propose a semi-decoupled approach to reduce the size of the total design space by orders of magnitude, yet without losing optimality. We first perform neural architecture search to obtain a small set of optimal architectures for one accelerator candidate. Importantly, this is also the set of (close-to-)optimal architectures for other accelerator designs based on the property that neural architectures' ranking orders in terms of inference latency and energy consumption on different accelerator designs are highly similar. Then, instead of considering all the possible architectures, we optimize the accelerator design only in combination with this small set of architectures, thus significantly reducing the total search cost. We validate our approach by conducting experiments on various architecture spaces for accelerator designs with different dataflows. Our results highlight that we can obtain the optimal design by only navigating over the reduced search space. 

   more » « less
3. Polymorphic Accelerators for Deep Neural Networks

   https://doi.org/10.1109/TC.2020.3048624  

   Azizimazreah, Arash; Chen, Lizhong (January 2021, IEEE Transactions on Computers)

   Deep neural networks (DNNs) come with many forms, such as convolutional neural networks, multilayer perceptron and recurrent neural networks, to meet diverse needs of machine learning applications. However, existing DNN accelerator designs, when used to execute multiple neural networks, suffer from underutilization of processing elements, heavy feature map traffic, and large area overhead. In this paper, we propose a novel approach, Polymorphic Accelerators, to address the flexibility issue fundamentally. We introduce the abstraction of logical accelerators to decouple the fixed mapping with physical resources. Three procedures are proposed that work collaboratively to reconfigure the accelerator for the current network that is being executed and to enable cross-layer data reuse among logical accelerators. Evaluation results show that the proposed approach achieves significant improvement in data reuse, inference latency and performance, e.g., 1.52x and 1.63x increase in throughput compared with state-of-the-art flexible dataflow approach and resource partitioning approach, respectively. This demonstrates the effectiveness and promise of polymorphic accelerator architecture. 

   more » « less
4. Reconfigurable Dataflow Optimization for Spatiotemporal Spiking Neural Computation on Systolic Array Accelerators

   Lee, Jeong-Jun; Li, Peng (October 2020, Proceedings IEEE International Conference on Computer Design)

   Spiking neural networks (SNNs) offer a promising biologically-plausible computing model and lend themselves to ultra-low-power event-driven processing on neuromorphic processors. Compared with the conventional artificial neural networks, SNNs are well-suited for processing complex spatiotemporal data. Despite its significance, dataflow optimization of spiking neural accelerator architectures has not been extensively studied. Recognizing the need for efficient processing of complex spatiotemporal data while considering the all-or-none nature of spiking activities, we propose holistic reconfigurable dataflow optimization for systolic array acceleration of spiking convolutional networks (S-CNNs). A novel scheme is introduced for parallel acceleration of computation across multiple time points, which further allows for systemic optimization of variable tiling for a large performance and efficiency gains. We show how variable tiling, in particular, the positioning of the temporal dimension, can be targeted to optimize data movement, throughput, and energy efficiency. Furthermore, we explore joint layer-dependent dataflow and accelerator hardware optimization to further boost performance and energy efficiency. To support systemic design space exploration, we develop an SNN dataflow simulator capable of analyzing the throughput and energy dissipation of systolic array accelerators for any targeted S-CNN while considering the inherent spatiotemporal characteristics of spiking neural computation. The proposed techniques deliver orders of magnitude of improvements on throughput, energy efficiency, and delay-energy product for accelerating deep Alexnet and VGG-16 SNNs. 

   more » « less
5. A Formalism of DNN Accelerator Flexibility

   https://doi.org/10.1145/3530907  

   Kao, Sheng-Chun; Kwon, Hyoukjun; Pellauer, Michael; Parashar, Angshuman; Krishna, Tushar (June 2022, Proceedings of the ACM on Measurement and Analysis of Computing Systems)

   The high efficiency of domain-specific hardware accelerators for machine learning (ML) has come fromspecialization, with the trade-off of less configurability/ flexibility. There is growing interest in developingflexible ML accelerators to make them future-proof to the rapid evolution of Deep Neural Networks (DNNs). However, the notion of accelerator flexibility has always been used in an informal manner, restricting computer architects from conducting systematic apples-to-apples design-space exploration (DSE) across trillions of choices. In this work, we formally define accelerator flexibility and show how it can be integrated for DSE. % flows. Specifically, we capture DNN accelerator flexibility across four axes: %the map-space of DNN accelerator along four flexibility axes: tiling, ordering, parallelization, and array shape. We categorize existing accelerators into 16 classes based on their axes of flexibility support, and define a precise quantification of the degree of flexibility of an accelerator across each axis. We leverage these to develop a novel flexibility-aware DSE framework. %It respects the difference of accelerator flexibility classes and degree of flexibility support in different accelerators, creating unique map-spaces. %and forms a unique map space for exploration. % We demonstrate how this can be used to perform first-of-their-kind evaluations, including an isolation study to identify the individual impact of the flexibility axes. We demonstrate that adding flexibility features to a hypothetical DNN accelerator designed in 2014 improves runtime on future (i.e., present-day) DNNs by 11.8x geomean. 

   more » « less