Modern high-performance computing (HPC) environments rely on hybrid storage systems (HSS) that combine multiple storage devices with diverse latency, bandwidth, endurance, and capacity characteristics to meet the performance, capacity, and cost requirements of data-intensive applications. The performance of an HSS highly depends on two key data-management policies: (1) data placement, which determines the most suitable storage device to store application data, and (2) data migration, which dynamically reorganizes previously-stored data across storage devices (i.e., prefetching hot data and evicting cold data) to sustain high HSS performance. These policies are tightly interdependent, and thus, improving one without considering the other leads to suboptimal HSS performance. Unfortunately, prior works optimize only one of the policies.

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$\sim 154\times$and$\sim 3113\times$, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.

Binary Neural Networks (BNNs) have demonstrated significant advantages in reducing computation and memory costs, all while maintaining acceptable accuracy on various image detection tasks. Thus, BNNs have the potential to support practical cognitive tasks on resource-constrained platforms, such as edge computing devices. To realize this, SRAM-based digital Computation-in-Memory (CIM) has gained growing attention as it overcomes the analog CIM architecture bottlenecks such as limited computing accuracy due to process variation, non-linearity, power and area-hungry Analog-to-Digital Converters (ADCs), etc. However, digital CIM architectures are highly dominated by power-hungry adder-trees, which can nullify the benefits of SRAM-based digital CIM. To address this issue, this paper proposes an adder free SRAM-based digital CIM, AFSRAM-CIM, for BNN acceleration. The proposed CIM architecture utilizes a multi-functional 10-T SRAM cell-based crossbar array and a new energy-efficient approach to perform the popcount operation. Simulation results using the MNIST dataset show that the proposed architecture maintains the state-of-the-art inference accuracy of 99.21% with only 11.86 fJ energy per operation. Moreover, AFSRAM-CIM achieves over$3\times$energy and$\approx 17\times$area savings when compared to the conventional digital CIM approaches.

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 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$\mathbf{446.8}\times$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$\mathbf{91.4}\times$compared to the SotA accelerator on GPU.

This paper investigates the potential of a compute-in-memory core based on optical Phase Change Materials (oPCMs) to speed up and reduce the energy consumption of the Matrix-Matrix-Multiplication operation. The paper also proposes a new data mapping for Binary Neural Networks (BNNs) tailored for our oPCM core. The preliminary results show a significant latency improvement irrespective of the evaluated network structure and size. The improvement varies from network to network and goes up to ~1053x.

Performing analysis on large graph datasets in an energy-efficient manner has posed a significant challenge; not only due to excessive data movements and poor locality, but also due to the non-optimal use of high sparsity of such datasets. The latter leads to a waste of resources as the computation is also performed on zero's operands which do not contribute to the final result. This paper designs a novel graph processing accelerator, SparseMEM, targeting sparse datasets by leveraging the computing-in-memory (CIM) concept; CIM is a promising solution to alleviate the overhead of data movement and the inherent poor locality of graph processing. The proposed solution stores the graph information in a compressed hierarchical format inside the memory and adjusts the workflow based on this new mapping. This vastly improves resource utilization, leading to higher energy and permanence efficiency. The experimental results demonstrate that SparseMEM outperforms a GPU-based platform and two state-of-the-art in-memory accelerators on speedup and energy efficiency by one and three orders of magnitude, respectively.

Basecalling, an essential step in many genome analysis studies, relies on large Deep Neural Network s (DNN s) to achieve high accuracy. Unfortunately, these DNN s are computationally slow and inefficient, leading to considerable delays and resource constraints in the sequence analysis process. A Computation-In-Memory (CIM) architecture using memristors can significantly accelerate the performance of DNN s. However, inherent device non-idealities and architectural limitations of such designs can greatly degrade the basecalling accuracy, which is critical for accurate genome analysis. To facilitate the adoption of memristor-based CIM designs for basecalling, it is important to (1) conduct a comprehensive analysis of potential CIM architectures and (2) develop effective strategies for mitigating the possible adverse effects of inherent device non-idealities and architectural limitations.

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.

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 $713 \times$ and $1.2 \times$, respectively, while simultaneously reducing energy consumption by an average of $1855 \times$ and $39.5 \times$. Across these workloads, pLUTo outperforms state-of-the-art PiM architectures by an average of $18.3 \times$. 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.

In recent years, we are witnessing a trend moving away from conventional computer architectures towards Computation-In-Memory (CIM) based on emerging memristor devices. This is due to the fact that the performance and energy efficiency of traditional computer architectures can no longer be increased at the same pace as before. The main barriers which limit the performance and energy improvement are the memory and power walls. Thus far, the main effort from researchers is toward enabling CIM as an accelerator for specific applications. Consequently, this current application-specific nature/approach has put less emphasis on the potential general-purpose applicability of CIM, i.e., merging several accelerators into one that is less than the sum of the parts. In this paper, we demonstrate the CIM concept using a broader and generalized model. Considering this model, the state-of-the-art CIM-based logic and arithmetic primitive functions, which can be the building blocks for complex functions, are investigated. Besides, we present potential applications of CIM which provides insights into the challenges and opportunities of a generic CIM system design. Finally, we highlight the future directions regarding the construction of CIM-based systems.

Aggressive memory density scaling causes modern DRAM devices to suffer from RowHammer, a phenomenon where rapidly activating (i.e., hammering) a DRAM row can cause bit-flips in physically-nearby rows. Recent studies demonstrate that modern DDR4/LPDDR4 DRAM chips, including chips previously marketed as RowHammer-safe, are even more vulnerable to RowHammer than older DDR3 DRAM chips. Many works show that attackers can exploit RowHammer bit-flips to reliably mount system-level attacks to escalate privilege and leak private data. Therefore, it is critical to ensure RowHammersafe operation on all DRAM-based systems as they become increasingly more vulnerable to RowHammer. Unfortunately, state-of-the-art RowHammer mitigation mechanisms face two major challenges. First, they incur increasingly higher performance and/or area overheads when applied to more vulnerable DRAM chips. Second, they require either closely-guarded proprietary information about the DRAM chips' physical circuit layouts or modifications to the DRAM chip design.In this paper, we show that it is possible to efficiently and scalably prevent RowHammer bit-flips without knowledge of or modification to DRAM internals. To this end, we introduce BlockHammer, a low-cost, effective, and easy-to-adopt RowHammer mitigation mechanism that prevents all RowHammer bit-flips while overcoming the two key challenges. BlockHammer selectively throttles memory accesses that could otherwise potentially cause RowHammer bit-flips. The key idea of BlockHammer is to (1) track row activation rates using area-efficient Bloom filters, and (2) use the tracking data to ensure that no row is ever activated rapidly enough to induce RowHammer bit-flips. By guaranteeing that no DRAM row ever experiences a RowHammer-unsafe activation rate, BlockHammer (1) makes it impossible for a RowHammer bit-flip to occur and (2) greatly reduces a RowHammer attack's impact on the performance of co-running benign applications. Our evaluations across a comprehensive range of 280 workloads show that, compared to the best of six state-of-the-art RowHammer mitigation mechanisms (all of which require knowledge of or modification to DRAM internals), BlockHammer provides (1) competitive performance and energy when the system is not under a RowHammer attack and (2) significantly better performance and energy when the system is under a RowHammer attack.

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 show that 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.

Conventional planar video streaming is the most popular application in mobile systems. The rapid growth of 360° video content and virtual reality (VR) devices is accelerating the adoption of VR video streaming. Unfortunately, video streaming consumes significant system energy due to high power consumption of major system components (e.g., DRAM, display interfaces, and display panel) involved in the video streaming process. For example, in conventional planar video streaming, the video decoder (in the processor) decodes video frames and stores them in the DRAM main memory before the display controller (in the processor) transfers decoded frames from DRAM to the display panel. This system architecture causes large amount of data movement to/from DRAM as well as high DRAM bandwidth usage. As a result, DRAM by itself consumes more than 30% of the video streaming energy.

MOTIVATION: We introduce SneakySnake, a highly parallel and highly accurate pre-alignment filter that remarkably reduces the need for computationally costly sequence alignment. The key idea of SneakySnake is to reduce the approximate string matching (ASM) problem to the single net routing (SNR) problem in VLSI chip layout. In the SNR problem, we are interested in finding the optimal path that connects two terminals with the least routing cost on a special grid layout that contains obstacles. The SneakySnake algorithm quickly solves the SNR problem and uses the found optimal path to decide whether or not performing sequence alignment is necessary. Reducing the ASM problem into SNR also makes SneakySnake efficient to implement on CPUs, GPUs and FPGAs. RESULTS: SneakySnake significantly improves the accuracy of pre-alignment filtering by up to four orders of magnitude compared to the state-of-the-art pre-alignment filters, Shouji, GateKeeper and SHD. For short sequences, SneakySnake accelerates Edlib (state-of-the-art implementation of Myers's bit-vector algorithm) and Parasail (state-of-the-art sequence aligner with a configurable scoring function), by up to 37.7× and 43.9× (>12× on average), respectively, with its CPU implementation, and by up to 413× and 689× (>400× on average), respectively, with FPGA and GPU acceleration. For long sequences, the CPU implementation of SneakySnake accelerates Parasail and KSW2 (sequence aligner of minimap2) by up to 979× (276.9× on average) and 91.7× (31.7× on average), respectively. As SneakySnake does not replace sequence alignment, users can still obtain all capabilities (e.g. configurable scoring functions) of the aligner of their choice, unlike existing acceleration efforts that sacrifice some aligner capabilities. AVAILABILITYAND IMPLEMENTATION: https://github.com/CMU-SAFARI/SneakySnake. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

DRAM is the prevalent main memory technology, but its long access latency can limit the performance of many workloads. Although prior works provide DRAM designs that reduce DRAM access latency, their reduced storage capacities hinder the performance of workloads that need large memory capacity. Because the capacity-latency trade-off is fixed at design time, previous works cannot achieve maximum performance under very different and dynamic workload demands.This paper proposes Capacity-Latency-Reconfigurable DRAM (CLR-DRAM), a new DRAM architecture that enables dynamic capacity-latency trade-off at low cost. CLR-DRAM allows dynamic reconfiguration of any DRAM row to switch between two operating modes: 1) max-capacity mode, where every DRAM cell operates individually to achieve approximately the same storage density as a density-optimized commodity DRAM chip and 2) high-performance mode, where two adjacent DRAM cells in a DRAM row and their sense amplifiers are coupled to operate as a single low-latency logical cell driven by a single logical sense amplifier.We implement CLR-DRAM by adding isolation transistors in each DRAM subarray. Our evaluations show that CLR-DRAM can improve system performance and DRAM energy consumption by 18.6% and 29.7% on average with four-core multiprogrammed workloads. We believe that CLR-DRAM opens new research directions for a system to adapt to the diverse and dynamically changing memory capacity and access latency demands of workloads.

Increasing single-cell DRAM error rates have pushed DRAM manufacturers to adopt on-die error-correction coding (ECC), which operates entirely within a DRAM chip to improve factory yield. The on-die ECC function and its effects on DRAM reliability are considered trade secrets, so only the manufacturer knows precisely how on-die ECC alters the externally-visible reliability characteristics. Consequently, on-die ECC obstructs third-party DRAM customers (e.g., test engineers, experimental researchers), who typically design, test, and validate systems based on these characteristicsTo give third parties insight into precisely how on-die ECC transforms DRAM error patterns during error correction, we introduce Bit-Exact ECC Recovery (BEER), a new methodology for determining the full DRAM on-die ECC function (i.e., its parity-check matrix) without hardware tools, prerequisite knowledge about the DRAM chip or on-die ECC mechanism, or access to ECC metadata (e.g., error syndromes, parity information). BEER exploits the key insight that non-intrusively inducing data-retention errors with carefully-crafted test pat-terns reveals behavior that is unique to a specific ECC functionWe use BEER to identify the ECC functions of 80 real LPDDR4 DRAM chips with on-die ECC from three major DRAM manufacturers. We evaluate BEER's correctness in simulation and performance on a real system to show that BEER is effective and practical across a wide range of on-die ECC functions. To demonstrate BEER's value, we propose and discuss several ways that third parties can use BEER to improve their design and testing practices. As a concrete example, we introduce and evaluate BEEP, the first error profiling method-ology that uses the known on-die ECC function to recover the number and bit-exact locations of unobservable raw bit errors responsible for observable post-correction errors.

Important workloads, such as machine learning and graph analytics applications, heavily involve sparse linear algebra operations. These operations use sparse matrix compression as an effective means to avoid storing zeros and performing unnecessary computation on zero elements. However, compression techniques like Compressed Sparse Row (CSR) that are widely used today introduce significant instruction overhead and expensive pointer-chasing operations to discover the positions of the non-zero elements. In this paper, we identify the discovery of the positions (i.e., indexing) of non-zero elements as a key bottleneck in sparse matrix-based workloads, which greatly reduces the benefits of compression.

The effectiveness of deep neural networks (DNN) in vision, speech, and language processing has prompted a tremendous demand for energy-efficient high-performance DNN inference systems. Due to the increasing memory intensity of most DNN workloads, main memory can dominate the system's energy consumption and stall time. One effective way to reduce the energy consumption and increase the performance of DNN inference systems is by using approximate memory, which operates with reduced supply voltage and reduced access latency parameters that violate standard specifications. Using approximate memory reduces reliability, leading to higher bit error rates. Fortunately, neural networks have an intrinsic capacity to tolerate increased bit errors. This can enable energy-efficient and high-performance neural network inference using approximate DRAM devices.