Applications of Binary Neural Networks (BNNs) are promising for embedded systems with hard constraints on energy and computing power. Contrary to conventional neural networks using floating-point datatypes, BNNs use binarized weights and activations to reduce memory and computation requirements. Memristors, emerging non-volatile memory devices, show great potential as a target implementation platform for BNNs by integrating storage and compute units. However, the efficiency of this hardware highly depends on how the network is mapped and executed on these devices. In this paper, we propose an efficient implementation of XNOR-based BNN to maximize parallelization. In this implementation, costly analog-to-digital converters are replaced with sense amplifiers with custom reference(s) to generate activation values. Besides, a novel mapping is introduced to minimize the overhead of data communication between convolution layers mapped to different memristor crossbars. This comes with extensive analytical and simulation-based analysis to evaluate the implication of different design choices considering the accuracy of the network. The results show that our approach achieves up to 5× energy-saving and 100× improvement in latency compared to baselines.

State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper, we first present a new data mapping, called TacitMap, suited for BNNs implemented based on a Computation-In-Memory (CIM) architecture. TacitMap maximizes the use of available parallelism, while CIM architecture eliminates the data movement overhead. We then propose a hardware accelerator based on optical phase change memory (oPCM) called EinsteinBarrier. Ein-steinBarrier incorporates TacitMap and adds an extra dimension for parallelism through wavelength division multiplexing, leading to extra latency reduction. The simulation results show that, compared to the SotA CIM baseline, TacitMap and EinsteinBarrier significantly improve execution time by up to ∼ 154× and ∼ 3113×, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.

DNA sequence alignment is a fundamental and computationally expensive operation in bioinformatics. Researchers have developed pre-alignment filters that effectively reduce the amount of data consumed by the alignment process by discarding locations that result in a poor match. However, the filtering operation itself is memory-intensive for which the conventional Von-Neumann architectures perform poorly. Therefore, recent designs advocate compute near memory (CNM) accelerators based on stacked DRAM and more exotic memory technologies such as racetrack memories (RTM). However, these designs only support small DNA reads of circa 100 nucleotides, referred to as short reads. This letter proposes a CNM system for handling both long and short reads. It introduces a novel data-placement solution that significantly increases parallelism and reduces overhead. Evaluation results show substantial reductions in execution time (1.32times1.32×) and energy consumption (50%), compared to the state-of-the-art.

Profile hidden Markov models (pHMMs) are widely employed in various bioinformatics applications to identify similarities between biological sequences, such as DNA or protein sequences. In pHMMs, sequences are represented as graph structures, where states and edges capture modifications (i.e., insertions, deletions, and substitutions) by assigning probabilities to them. These probabilities are subsequently used to compute the similarity score between a sequence and a pHMM graph. The Baum-Welch algorithm, a prevalent and highly accurate method, utilizes these probabilities to optimize and compute similarity scores. Accurate computation of these probabilities is essential for the correct identification of sequence similarities. However, the Baum-Welch algorithm is computationally intensive, and existing solutions offer either software-only or hardware-only approaches with fixed pHMM designs. When we analyze state-of-the-art works, we identify an urgent need for a flexible, high-performance, and energy-efficient hardware-software co-design to address the major inefficiencies in the Baum-Welch algorithm for pHMMs. We introduce ApHMM, the first flexible acceleration framework designed to significantly reduce both computational and energy overheads associated with the Baum-Welch algorithm for pHMMs. ApHMM employs hardware-software co-design to tackle the major inefficiencies in the Baum-Welch algorithm by (1) designing flexible hardware to accommodate various pHMM designs, (2) exploiting predictable data dependency patterns through on-chip memory with memoization techniques, (3) rapidly filtering out unnecessary computations using a hardware-based filter, and (4) minimizing redundant computations. ApHMM achieves substantial speedups of 15.55×–260.03×, 1.83×–5.34×, and 27.97× when compared to CPU, GPU, and FPGA implementations of the Baum-Welch algorithm, respectively. ApHMM outperforms state-of-the-art CPU implementations in three key bioinformatics applications: (1) error correction, (2) protein family search, and (3) multiple sequence alignment, by 1.29×–59.94×, 1.03×–1.75×, and 1.03×–1.95×, respectively, while improving their energy efficiency by 64.24×–115.46×, 1.75×, and 1.96×.

The vast potential of memristor-based computation-in-memory (CIM) engines has mainly triggered the mapping of best-suited applications. Nevertheless, with additional support, existing applications can also benefit from CIM. In particular, this paper proposes an energy and area-efficient CIM-based methodology to perform arithmetic signed matrix multiplications. Our approach combines a) the mapping of the signed operands on the 1T1R crossbar, and b) the augmentation of the periphery with customized circuits to support the execution of shift and accumulate needed for the arithmetic operations. The operand mapping is performed without the need for sign extension; hence, reducing the required memory size. To demonstrate the superiority of our scheme as compared with the state-of-the-art, simulations are performed for different case studies including a neural network and two kernels which are taken from the Polybench/C benchmark suite. The results show that our approach achieves up to 8× energy-saving and 3× area-saving compared with other CIM-based prior works.

The high execution time of DNA sequence alignment negatively affects many genomic studies that rely on sequence alignment results. Pre-alignment filtering was introduced as a step before alignment to reduce the execution time of short-read sequence alignment greatly. With its success, i.e., achieving high accuracy and thus removing unnecessary alignments, the filtering itself now constitutes the larger portion of the execution time. A significant contributing factor entails the movement of sequences from the memory to the processing units, while a majority will filter out as they do not result in an acceptable alignment. State-of-the-art (SotA) pre-alignment filtering accelerators suffer from the same overhead for data movements. Furthermore, these accelerators lack support for future pre-alignment filtering algorithms using the same operations and underlying hardware. This paper addresses these shortcomings by introducing SieveMem. SieveMem is an architecture that exploits the Computation-in-Memory paradigm with memristive-based devices to support shared kernels of pre-alignment filters and algorithms inside the memory (i.e., preventing data movements). SieveMem architecture also provides support for future algorithms. SieveMem supports more than 47.6% of shared operations among all top 5 SotA filters. Moreover, SieveMem includes a hardware-friendly pre-alignment filtering algorithm called BandedKrait, inspired by a combination of mentioned kernels. Our evaluations show that SieveMem provides up to 331.1 x and 446.8 × improvement in the execution time of the two most-common kernels. Our evaluations also show that BandedKrait provides accuracy at the SotA level. Using BandedKrait on SieveMem, a design we call Mem-BandedKrait, one can improve the execution time of end-to-end sequence alignment irrespective of the dataset, which can go up to 91.4 × compared to the SotA accelerator on GPU.

Data movement between the main memory and the processor is a key contributor to execution time and energy consumption in memory-intensive applications. This data movement bottleneck can be alleviated using Processing-in-Memory (PiM). One category of PiM is Processing-using-Memory (PuM), in which computation takes place inside the memory array by exploiting intrinsic analog properties of the memory device. PuM yields high performance and energy efficiency, but existing PuM techniques support a limited range of operations. As a result, current PuM architectures cannot efficiently perform some complex operations (e.g., multiplication, division, exponentiation) without large increases in chip area and design complexity. To overcome these limitations of existing PuM architectures, we introduce pLUTo (processing-using-memory with lookup table (LUT) operations), a DRAM-based PuM architecture that leverages the high storage density of DRAM to enable the massively parallel storing and querying of lookup tables (LUTs). The key idea of pLUTo is to replace complex operations with low-cost, bulk memory reads (i.e., LUT queries) instead of relying on complex extra logic. We evaluate pLUTo across 11 real-world workloads that showcase the limitations of prior PuM approaches and show that our solution outperforms optimized CPU and GPU base-lines by an average of 713 × and 1.2 ×, respectively, while simultaneously reducing energy consumption by an average of 1855 × and 39.5 ×. Across these workloads, pLUTo outperforms state-of-the-art PiM architectures by an average of 18.3 ×. We also show that different versions of pLUTo provide different levels of flexibility and performance at different additional DRAM area overheads (between 10.2% and 23.1%). pLUTo's source code and all scripts required to reproduce the results of this paper are openly and fully available at https://github.com/CMU-SAFARI/pLUTo.

Spin-transfer torque magnetic random access memory (STT-MRAM) based computation-in-memory (CIM) architectures have shown great prospects for an energy-efficient computing. However, device variations and non-idealities narrow down the sensing margin that severely impacts the computing accuracy. In this work, we propose an adaptive referencing mechanism to improve the sensing margin of a CIM architecture for boolean binary logic (BBL) operations. We generate reference signals using multiple STT-MRAM devices and place them strategically into the array such that these signals can address the variations and trace the wire parasitics effectively. We have demonstrated this behavior using an STT-MRAM model, which is calibrated using 1Mbit characterized array. Results show that our proposed architecture for binary neural networks (BNN) achieves up to 17.8 TOPS/W on the MNIST dataset and 130× performance improvement for the text encryption compared to the software implementation on Intel Haswell processor.

State-of-the-art taxonomic profilers that comprise the first step in larger-context metagenomic studies have proven to be computationally intensive, i.e., while accurate, they come at the cost of high latency and energy consumption. Table Lookup operation is a primary bottleneck of today's profilers. In this paper, we first propose TL-PIM, a hardware accelerator based on the processing-in-memory (PIM) paradigm to accelerate Table Lookup. TL-PIM leverages the in-memory compute capability of emerging memory technologies along with intelligent data mapping. Then, we integrate TL-PIM into Kraken2, a state-of-the-art metagenomic profiler, and build an HW/SW co-designed profiler, called KrakenOnMem. Results from a silicon-based prototype of our emerging memory validate the design and required operations on a smaller scale. Our large-scale calibrated simulations show that KrakenOnMem can provide an average of 61.3% speedup compared to original Kraken2 for end-to-end profiling. Additionally, our design improves the energy consumption by orders of magnitude compared to the original Kraken2 while incurring a negligible area overhead.

Past research has proposed numerous hardware prefetching techniques, most of which rely on exploiting one specific type of program context information (e.g., program counter, cacheline address, or delta between cacheline addresses) to predict future memory accesses. These techniques either completely neglect a prefetcher's undesirable effects (e.g., memory bandwidth usage) on the overall system, or incorporate system-level feedback as an afterthought to a system-unaware prefetch algorithm.We showthat prior prefetchers often lose their performance benefit over a wide range of workloads and system configurations due to their inherent inability to take multiple different types of program context and system-level feedback information into account while prefetching. In this paper, we make a case for designing a holistic prefetch algorithm that learns to prefetch using multiple different types of program context and system-level feedback information inherent to its design. To this end, we propose Pythia, which formulates the prefetcher as a reinforcement learning agent. For every demand request, Pythia observes multiple different types of program context information to make a prefetch decision. For every prefetch decision, Pythia receives a numerical reward that evaluates prefetch quality under the current memory bandwidth usage. Pythia uses this reward to reinforce the correlation between program context information and prefetch decision to generate highly accurate, timely, and systemaware prefetch requests in the future. Our extensive evaluations using simulation and hardware synthesis show that Pythia outperforms two state-of-the-art prefetchers (MLOP and Bingo) by 3.4% and 3.8% in single-core, 7.7% and 9.6% in twelve-core, and 16.9% and 20.2% in bandwidth-constrained core configurations, while incurring only 1.03% area overhead over a desktop-class processor and no software changes in workloads. The source code of Pythia can be freely downloaded from https://github.com/CMU-SAFARI/Pythia.