More Like this

1. Efficient Memory Integration: MRAM-SRAM Hybrid Accelerator for Sparse On-Device Learning

   https://doi.org/10.1145/3649329.3657390  

   Zhang, Fan; Sridharan, Amitesh; Tsai, Wilman; Chen, Yiran; Wang, Shan X; Fan, Deliang (June 2024, ACM)

   With the prosperous development of Deep Neural Network (DNNs), numerous Process-In-Memory (PIM) designs have emerged to accelerate DNN models with exceptional throughput and energy-efficiency. PIM accelerators based on Non-Volatile Memory (NVM) or volatile memory offer distinct advantages for computational efficiency and performance. NVM based PIM accelerators, demonstrated success in DNN inference, face limitations in on-device learning due to high write energy, latency, and instability. Conversely, fast volatile memories, like SRAM, offer rapid read/write operations for DNN training, but suffer from significant leakage currents and large memory footprints. In this paper, for the first time, we present a fully-digital sparse processing in hybrid NVM-SRAM design, synergistically combines the strengths of NVM and SRAM, tailored for on-device continual learning. Our designed NVM and SRAM based PIM circuit macros could support both storage and processing of N:M structured sparsity pattern, significantly improving the storage and computing efficiency. Exhaustive experiments demonstrate that our hybrid system effectively reduces area and power consumption while maintaining high accuracy, offering a scalable and versatile solution for on-device continual learning. 

   more » « less
2. Deep Neural Network Acceleration in Non-Volatile Memory: A Digital Approach

   https://doi.org/10.1109/NANOARCH47378.2019.181297  

   Angizi, Shaahin; Fan, Deliang (July 2019, 10.1109/NANOARCH47378.2019.181297)

   Latest algorithmic development has brought competitive classification accuracy for neural networks despite constraining the network parameters to ternary or binary representations. These findings show significant optimization opportunities to replace computationally-intensive convolution operations (based on multiplication) with more efficient and less complex operations such as addition. In hardware implementation domain, processing-in-memory architecture is becoming a promising solution to alleviate enormous energy-hungry data communication between memory and processing units, bringing considerable improvement for system performance and energy efficiency while running such large networks. In this paper, we review several of our recent works regarding Processing-in-Memory (PIM) accelerator based on Magnetic Random Access Memory computational sub-arrays to accelerate the inference mode of quantized neural networks using digital non-volatile memory rather than using analog crossbar operation. In this way, we investigate the performance of two distinct in-memory addition schemes compared to other digital methods based on processing-in-DRAM/GPU/ASIC design to tackle DNN power and memory wall bottleneck. 

   more » « less
3. Accelerating Deep Neural Networks in Processing-in-Memory Platforms: Analog or Digital Approach?

   https://doi.org/10.1109/ISVLSI.2019.00044  

   Angizi, Shaahin; He, Zhezhi; Reis, Dayane; Hu, Xiaobo Sharon; Tsai, Wilman; Lin, Shy Jay; Fan, Deliang (July 2019, 2019 IEEE Computer Society Annual Symposium on VLSI (ISVLSI))

   Nowadays, research topics on AI accelerator designs have attracted great interest, where accelerating Deep Neural Network (DNN) using Processing-in-Memory (PIM) platforms is an actively-explored direction with great potential. PIM platforms, which simultaneously aims to address power- and memory-wall bottlenecks, have shown orders of performance enhancement in comparison to the conventional computing platforms with Von-Neumann architecture. As one direction of accelerating DNN in PIM, resistive memory array (aka. crossbar) has drawn great research interest owing to its analog current-mode weighted summation operation which intrinsically matches the dominant Multiplication-and-Accumulation (MAC) operation in DNN, making it one of the most promising candidates. An alternative direction for PIM-based DNN acceleration is through bulk bit-wise logic operations directly performed on the content in digital memories. Thanks to the high fault-tolerant characteristic of DNN, the latest algorithmic progression successfully quantized DNN parameters to low bit-width representations, while maintaining competitive accuracy levels. Such DNN quantization techniques essentially convert MAC operation to much simpler addition/subtraction or comparison operations, which can be performed by bulk bit-wise logic operations in a highly parallel fashion. In this paper, we build a comprehensive evaluation framework to quantitatively compare and analyze aforementioned PIM based analog and digital approaches for DNN acceleration. 

   more » « less
4. Improving the Robustness and Efficiency of PIM-Based Architecture by SW/HW Co-Design

   https://doi.org/10.1145/3566097.3568358  

   Yang, Xiaoxuan; Li, Shiyu; Zheng, Qilin; Chen, Yiran (January 2023, 2023 28th Asia and South Pacific Design Automation Conference (ASP-DAC))

   Processing-in-memory (PIM) based architecture shows great potential to process several emerging artificial intelligence workloads, including vision and language models. Cross-layer optimizations could bridge the gap between computing density and the available resources by reducing the computation and memory cost of the model and improving the model’s robustness against non-ideal hardware effects. We first introduce several hardware-aware training methods to improve the model robustness to the PIM device’s nonideal effects, including stuck-at-fault, process variation, and thermal noise. Then, we further demonstrate a software/hardware (SW/HW) co-design methodology to efficiently process the state-of-the-art attention-based model on PIM-based architecture by performing sparsity exploration for the attention-based model and circuit architecture co-design to support the sparse processing. 

   more » « less
5. PIM-trie: A Skew-resistant Trie for Processing-in-Memory

   https://doi.org/10.1145/3558481.3591070  

   Kang, Hongbo; Zhao, Yiwei; Blelloch, Guy E.; Dhulipala, Laxman; Gu, Yan; McGuffey, Charles; Gibbons, Phillip B. (June 2023, ACM)

   Memory latency and bandwidth are significant bottlenecks in designing in-memory indexes. Processing-in-memory (PIM), an emerging hardware design approach, alleviates this problem by embedding processors in memory modules, enabling low-latency memory access whose aggregated bandwidth scales linearly with the number of PIM modules. Despite recent work in balanced comparison-based indexes on PIM systems, building efficient tries for PIMs remains an open challenge due to tries' inherently unbalanced shape. This paper presents the PIM-trie, the first batch-parallel radix-based index for PIM systems that provides load balance and low communication under adversary-controlled workloads. We introduce trie matching-matching a query trie of a batch against the compressed data trie-as a key building block for PIM-friendly index operations. Our algorithm combines (i) hash-based comparisons for coarse-grained work distribution/elimination and (ii) bit-by-bit comparisons for fine-grained matching. Combined with other techniques (meta-block decomposition, selective recursive replication, differentiated verification), PIM-trie supports LongestCommonPrefix, Insert, and Delete in O(logP) communication rounds per batch and O(l/w) communication volume per string, where P is the number of PIM modules, l is the string length in bits, and w is the machine word size. Moreover, work and communication are load-balanced among modules whp, even under worst-case skew. 

   more » « less