arXiv subjects

Onur Mutlu

Publications and source records attributed to Onur Mutlu.

At least 163 records · Page 9Linked to original sources

arXiv2023

Benchmarking Memory-Centric Computing Systems: Analysis of Real Processing-in-Memory Hardware

Many modern workloads such as neural network inference and graph processing are fundamentally memory-bound. For such workloads, data movement between memory and CPU cores imposes a significant overhead in terms of both latency and energy. A major reason is that this communication happens through a narrow bus with high latency and limited bandwidth, and the low data reuse in memory-bound workloads is insufficient to amortize the cost of memory access. Fundamentally addressing this data movement bottleneck requires a paradigm where the memory system assumes an active role in computing by integrating processing capabilities. This paradigm is known as processing-in-memory (PIM). Recent research explores different forms of PIM architectures, motivated by the emergence of new technologies that integrate memory with a logic layer, where processing elements can be easily placed. Past works evaluate these architectures in simulation or, at best, with simplified hardware prototypes. In contrast, the UPMEM company has designed and manufactured the first publicly-available real-world PIM architecture. The UPMEM PIM architecture combines traditional DRAM memory arrays with general-purpose in-order cores, called DRAM Processing Units (DPUs), integrated in the same chip. This paper presents key takeaways from the first comprehensive analysis of the first publicly-available real-world PIM architecture. We provide four key takeaways about the UPMEM PIM architecture, which stem from our study. More insights about suitability of different workloads to the PIM system, programming recommendations for software designers, and suggestions and hints for hardware and architecture designers of future PIM systems are available in arXiv:2105.03814

cs.AR↗

arXiv2023

A Framework for High-throughput Sequence Alignment using Real Processing-in-Memory Systems

Sequence alignment is a memory bound computation whose performance in modern systems is limited by the memory bandwidth bottleneck. Processing-in-memory architectures alleviate this bottleneck by providing the memory with computing competencies. We propose Alignment-in-Memory (AIM), a framework for high-throughput sequence alignment using processing-in-memory, and evaluate it on UPMEM, the first publicly-available general-purpose programmable processing-in-memory system. Our evaluation shows that a real processing-in-memory system can substantially outperform server-grade multi-threaded CPU systems running at full-scale when performing sequence alignment for a variety of algorithms, read lengths, and edit distance thresholds. We hope that our findings inspire more work on creating and accelerating bioinformatics algorithms for such real processing-in-memory systems. Our code is available at https://github.com/safaad/aim.

cs.AR↗

arXiv2023

Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

Neural networks (NNs) are growing in importance and complexity. A neural network's performance (and energy efficiency) can be bound either by computation or memory resources. The processing-in-memory (PIM) paradigm, where computation is placed near or within memory arrays, is a viable solution to accelerate memory-bound NNs. However, PIM architectures vary in form, where different PIM approaches lead to different trade-offs. Our goal is to analyze, discuss, and contrast DRAM-based PIM architectures for NN performance and energy efficiency. To do so, we analyze three state-of-the-art PIM architectures: (1) UPMEM, which integrates processors and DRAM arrays into a single 2D chip; (2) Mensa, a 3D-stack-based PIM architecture tailored for edge devices; and (3) SIMDRAM, which uses the analog principles of DRAM to execute bit-serial operations. Our analysis reveals that PIM greatly benefits memory-bound NNs: (1) UPMEM provides 23x the performance of a high-end GPU when the GPU requires memory oversubscription for a general matrix-vector multiplication kernel; (2) Mensa improves energy efficiency and throughput by 3.0x and 3.1x over the Google Edge TPU for 24 Google edge NN models; and (3) SIMDRAM outperforms a CPU/GPU by 16.7x/1.4x for three binary NNs. We conclude that the ideal PIM architecture for NN models depends on a model's distinct attributes, due to the inherent architectural design choices.

cs.AR↗

arXiv2023

Fundamentally Understanding and Solving RowHammer

We provide an overview of recent developments and future directions in the RowHammer vulnerability that plagues modern DRAM (Dynamic Random Memory Access) chips, which are used in almost all computing systems as main memory. RowHammer is the phenomenon in which repeatedly accessing a row in a real DRAM chip causes bitflips (i.e., data corruption) in physically nearby rows. This phenomenon leads to a serious and widespread system security vulnerability, as many works since the original RowHammer paper in 2014 have shown. Recent analysis of the RowHammer phenomenon reveals that the problem is getting much worse as DRAM technology scaling continues: newer DRAM chips are fundamentally more vulnerable to RowHammer at the device and circuit levels. Deeper analysis of RowHammer shows that there are many dimensions to the problem as the vulnerability is sensitive to many variables, including environmental conditions (temperature \& voltage), process variation, stored data patterns, as well as memory access patterns and memory control policies. As such, it has proven difficult to devise fully-secure and very efficient (i.e., low-overhead in performance, energy, area) protection mechanisms against RowHammer and attempts made by DRAM manufacturers have been shown to lack security guarantees. After reviewing various recent developments in exploiting, understanding, and mitigating RowHammer, we discuss future directions that we believe are critical for solving the RowHammer problem. We argue for two major directions to amplify research and development efforts in: 1) building a much deeper understanding of the problem and its many dimensions, in both cutting-edge DRAM chips and computing systems deployed in the field, and 2) the design and development of extremely efficient and fully-secure solutions via system-memory cooperation.

cs.CR↗

arXiv2023

MetaSys: A Practical Open-Source Metadata Management System to Implement and Evaluate Cross-Layer Optimizations

This paper introduces the first open-source FPGA-based infrastructure, MetaSys, with a prototype in a RISC-V core, to enable the rapid implementation and evaluation of a wide range of cross-layer techniques in real hardware. Hardware-software cooperative techniques are powerful approaches to improve the performance, quality of service, and security of general-purpose processors. They are however typically challenging to rapidly implement and evaluate in real hardware as they require full-stack changes to the hardware, OS, system software, and instruction-set architecture (ISA). MetaSys implements a rich hardware-software interface and lightweight metadata support that can be used as a common basis to rapidly implement and evaluate new cross-layer techniques. We demonstrate MetaSys's versatility and ease-of-use by implementing and evaluating three cross-layer techniques for: (i) prefetching for graph analytics; (ii) bounds checking in memory unsafe languages, and (iii) return address protection in stack frames; each technique only requiring ~100 lines of Chisel code over MetaSys. Using MetaSys, we perform the first detailed experimental study to quantify the performance overheads of using a single metadata management system to enable multiple cross-layer optimizations in CPUs. We identify the key sources of bottlenecks and system inefficiency of a general metadata management system. We design MetaSys to minimize these inefficiencies and provide increased versatility compared to previously-proposed metadata systems. Using three use cases and a detailed characterization, we demonstrate that a common metadata management system can be used to efficiently support diverse cross-layer techniques in CPUs.

cs.AR↗

arXiv2023

High-Performance and Scalable Agent-Based Simulation with BioDynaMo

Agent-based modeling plays an essential role in gaining insights into biology, sociology, economics, and other fields. However, many existing agent-based simulation platforms are not suitable for large-scale studies due to the low performance of the underlying simulation engines. To overcome this limitation, we present a novel high-performance simulation engine. We identify three key challenges for which we present the following solutions. First, to maximize parallelization, we present an optimized grid to search for neighbors and parallelize the merging of thread-local results. Second, we reduce the memory access latency with a NUMA-aware agent iterator, agent sorting with a space-filling curve, and a custom heap memory allocator. Third, we present a mechanism to omit the collision force calculation under certain conditions. Our evaluation shows an order of magnitude improvement over Biocellion, three orders of magnitude speedup over Cortex3D and NetLogo, and the ability to simulate 1.72 billion agents on a single server. Supplementary Materials, including instructions to reproduce the results, are available at: https://doi.org/10.5281/zenodo.6463816

cs.DC↗

arXiv2023

RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

In recent years, Convolutional Neural Networks (CNNs) have become the standard class of deep neural network for image processing, classification and segmentation tasks. However, the large strides in accuracy obtained by CNNs have been derived from increasing the complexity of network topologies, which incurs sizeable performance and energy penalties in the training and inference of CNNs. Many recent works have validated the effectiveness of parameter quantization, which consists in reducing the bit width of the network's parameters, to enable the attainment of considerable performance and energy efficiency gains without significantly compromising accuracy. However, it is difficult to compare the relative effectiveness of different quantization methods. To address this problem, we introduce RedBit, an open-source framework that provides a transparent, extensible and easy-to-use interface to evaluate the effectiveness of different algorithms and parameter configurations on network accuracy. We use RedBit to perform a comprehensive survey of five state-of-the-art quantization methods applied to the MNIST, CIFAR-10 and ImageNet datasets. We evaluate a total of 2300 individual bit width combinations, independently tuning the width of the network's weight and input activation parameters, from 32 bits down to 1 bit (e.g., 8/8, 2/2, 1/32, 1/1, for weights/activations). Upwards of 20000 hours of computing time in a pool of state-of-the-art GPUs were used to generate all the results in this paper. For 1-bit quantization, the accuracy losses for the MNIST, CIFAR-10 and ImageNet datasets range between [0.26%, 0.79%], [9.74%, 32.96%] and [10.86%, 47.36%] top-1, respectively. We actively encourage the reader to download the source code and experiment with RedBit, and to submit their own observed results to our public repository, available at https://github.com/IT-Coimbra/RedBit.

cs.LG↗

arXiv2022

ALP: Alleviating CPU-Memory Data Movement Overheads in Memory-Centric Systems

Partitioning applications between NDP and host CPU cores causes inter-segment data movement overhead, which is caused by moving data generated from one segment (e.g., instructions, functions) and used in consecutive segments. Prior works take two approaches to this problem. The first class of works maps segments to NDP or host cores based on the properties of each segment, neglecting the inter-segment data movement overhead. The second class of works partitions applications based on the overall memory bandwidth saving of each segment, and does not offload each segment to the best-fitting core if they incur high inter-segment data movement. We show that 1) mapping each segment to its best-fitting core ideally can provide substantial benefits, and 2) the inter-segment data movement reduces this benefit significantly. To this end, we introduce ALP, a new programmer-transparent technique to leverage the performance benefits of NDP by alleviating the inter-segment data movement overhead between host and memory and enabling efficient partitioning of applications. ALP alleviates the inter-segment data movement overhead by proactively and accurately transferring the required data between the segments. This is based on the key observation that the instructions that generate the inter-segment data stay the same across different executions of a program on different inputs. ALP uses a compiler pass to identify these instructions and uses specialized hardware to transfer data between the host and NDP cores at runtime. ALP efficiently maps application segments to either host or NDP considering 1) the properties of each segment, 2) the inter-segment data movement overhead, and 3) whether this overhead can be alleviated in a timely manner. We evaluate ALP across a wide range of workloads and show on average 54.3% and 45.4% speedup compared to only-host CPU or only-NDP executions, respectively.

cs.AR↗

arXiv2022

ProbGraph: High-Performance and High-Accuracy Graph Mining with Probabilistic Set Representations

Important graph mining problems such as Clustering are computationally demanding. To significantly accelerate these problems, we propose ProbGraph: a graph representation that enables simple and fast approximate parallel graph mining with strong theoretical guarantees on work, depth, and result accuracy. The key idea is to represent sets of vertices using probabilistic set representations such as Bloom filters. These representations are much faster to process than the original vertex sets thanks to vectorizability and small size. We use these representations as building blocks in important parallel graph mining algorithms such as Clique Counting or Clustering. When enhanced with ProbGraph, these algorithms significantly outperform tuned parallel exact baselines (up to nearly 50x on 32 cores) while ensuring accuracy of more than 90% for many input graph datasets. Our novel bounds and algorithms based on probabilistic set representations with desirable statistical properties are of separate interest for the data analytics community.

cs.DC↗

arXiv2022

TuRaN: True Random Number Generation Using Supply Voltage Underscaling in SRAMs

Prior works propose SRAM-based TRNGs that extract entropy from SRAM arrays. SRAM arrays are widely used in a majority of specialized or general-purpose chips that perform the computation to store data inside the chip. Thus, SRAM-based TRNGs present a low-cost alternative to dedicated hardware TRNGs. However, existing SRAM-based TRNGs suffer from 1) low TRNG throughput, 2) high energy consumption, 3) high TRNG latency, and 4) the inability to generate true random numbers continuously, which limits the application space of SRAM-based TRNGs. Our goal in this paper is to design an SRAM-based TRNG that overcomes these four key limitations and thus, extends the application space of SRAM-based TRNGs. To this end, we propose TuRaN, a new high-throughput, energy-efficient, and low-latency SRAM-based TRNG that can sustain continuous operation. TuRaN leverages the key observation that accessing SRAM cells results in random access failures when the supply voltage is reduced below the manufacturer-recommended supply voltage. TuRaN generates random numbers at high throughput by repeatedly accessing SRAM cells with reduced supply voltage and post-processing the resulting random faults using the SHA-256 hash function. To demonstrate the feasibility of TuRaN, we conduct SPICE simulations on different process nodes and analyze the potential of access failure for use as an entropy source. We verify and support our simulation results by conducting real-world experiments on two commercial off-the-shelf FPGA boards. We evaluate the quality of the random numbers generated by TuRaN using the widely-adopted NIST standard randomness tests and observe that TuRaN passes all tests. TuRaN generates true random numbers with (i) an average (maximum) throughput of 1.6Gbps (1.812Gbps), (ii) 0.11nJ/bit energy consumption, and (iii) 278.46us latency.

cs.AR↗

arXiv2022

NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

Resistive Random-Access Memory (RRAM) is well-suited to accelerate neural network (NN) workloads as RRAM-based Processing-in-Memory (PIM) architectures natively support highly-parallel multiply-accumulate (MAC) operations that form the backbone of most NN workloads. Unfortunately, NN workloads such as transformers require support for non-MAC operations (e.g., softmax) that RRAM cannot provide natively. Consequently, state-of-the-art works either integrate additional digital logic circuits to support the non-MAC operations or offload the non-MAC operations to CPU/GPU, resulting in significant performance and energy efficiency overheads due to data movement. In this work, we propose NEON, a novel compiler optimization to enable the end-to-end execution of the NN workload in RRAM. The key idea of NEON is to transform each non-MAC operation into a lightweight yet highly-accurate neural network. Utilizing neural networks to approximate the non-MAC operations provides two advantages: 1) We can exploit the key strength of RRAM, i.e., highly-parallel MAC operation, to flexibly and efficiently execute non-MAC operations in memory. 2) We can simplify RRAM's microarchitecture by eliminating the additional digital logic circuits while reducing the data movement overheads. Acceleration of the non-MAC operations in memory enables NEON to achieve a 2.28x speedup compared to an idealized digital logic-based RRAM. We analyze the trade-offs associated with the transformation and demonstrate feasible use cases for NEON across different substrates.

cs.AR↗

arXiv2022

Uncovering In-DRAM RowHammer Protection Mechanisms: A New Methodology, Custom RowHammer Patterns, and Implications

The RowHammer vulnerability in DRAM is a critical threat to system security. To protect against RowHammer, vendors commit to security-through-obscurity: modern DRAM chips rely on undocumented, proprietary, on-die mitigations, commonly known as Target Row Refresh (TRR). At a high level, TRR detects and refreshes potential RowHammer-victim rows, but its exact implementations are not openly disclosed. Security guarantees of TRR mechanisms cannot be easily studied due to their proprietary nature. To assess the security guarantees of recent DRAM chips, we present Uncovering TRR (U-TRR), an experimental methodology to analyze in-DRAM TRR implementations. U-TRR is based on the new observation that data retention failures in DRAM enable a side channel that leaks information on how TRR refreshes potential victim rows. U-TRR allows us to (i) understand how logical DRAM rows are laid out physically in silicon; (ii) study undocumented on-die TRR mechanisms; and (iii) combine (i) and (ii) to evaluate the RowHammer security guarantees of modern DRAM chips. We show how U-TRR allows us to craft RowHammer access patterns that successfully circumvent the TRR mechanisms employed in 45 DRAM modules of the three major DRAM vendors. We find that the DRAM modules we analyze are vulnerable to RowHammer, having bit flips in up to 99.9% of all DRAM rows. We make U-TRR source code openly and freely available at [106].

cs.CR↗

arXiv2022

LEAPER: Fast and Accurate FPGA-based System Performance Prediction via Transfer Learning

Machine learning has recently gained traction as a way to overcome the slow accelerator generation and implementation process on an FPGA. It can be used to build performance and resource usage models that enable fast early-stage design space exploration. First, training requires large amounts of data (features extracted from design synthesis and implementation tools), which is cost-inefficient because of the time-consuming accelerator design and implementation process. Second, a model trained for a specific environment cannot predict performance or resource usage for a new, unknown environment. In a cloud system, renting a platform for data collection to build an ML model can significantly increase the total-cost-ownership (TCO) of a system. Third, ML-based models trained using a limited number of samples are prone to overfitting. To overcome these limitations, we propose LEAPER, a transfer learning-based approach for prediction of performance and resource usage in FPGA-based systems. The key idea of LEAPER is to transfer an ML-based performance and resource usage model trained for a low-end edge environment to a new, high-end cloud environment to provide fast and accurate predictions for accelerator implementation. Experimental results show that LEAPER (1) provides, on average across six workloads and five FPGAs, 85% accuracy when we use our transferred model for prediction in a cloud environment with 5-shot learning and (2) reduces design-space exploration time for accelerator implementation on an FPGA by 10x, from days to only a few hours.

cs.AR↗

arXiv2022

HiRA: Hidden Row Activation for Reducing Refresh Latency of Off-the-Shelf DRAM Chips

DRAM is the building block of modern main memory systems. DRAM cells must be periodically refreshed to prevent data loss. Refresh operations degrade system performance by interfering with memory accesses. As DRAM chip density increases with technology node scaling, refresh operations also increase because: 1) the number of DRAM rows in a chip increases; and 2) DRAM cells need additional refresh operations to mitigate bit failures caused by RowHammer, a failure mechanism that becomes worse with technology node scaling. Thus, it is critical to enable refresh operations at low performance overhead. To this end, we propose a new operation, Hidden Row Activation (HiRA), and the HiRA Memory Controller (HiRA-MC). HiRA hides a refresh operation's latency by refreshing a row concurrently with accessing or refreshing another row within the same bank. Unlike prior works, HiRA achieves this parallelism without any modifications to off-the-shelf DRAM chips. To do so, it leverages the new observation that two rows in the same bank can be activated without data loss if the rows are connected to different charge restoration circuitry. We experimentally demonstrate on 56% real off-the-shelf DRAM chips that HiRA can reliably parallelize a DRAM row's refresh operation with refresh or activation of any of the 32% of the rows within the same bank. By doing so, HiRA reduces the overall latency of two refresh operations by 51.4%. HiRA-MC modifies the memory request scheduler to perform HiRA when a refresh operation can be performed concurrently with a memory access or another refresh. Our system-level evaluations show that HiRA-MC increases system performance by 12.6% and 3.73x as it reduces the performance degradation due to periodic refreshes and refreshes for RowHammer protection (preventive refreshes), respectively, for future DRAM chips with increased density and RowHammer vulnerability.

cs.AR↗

arXiv2022

Flash-Cosmos: In-Flash Bulk Bitwise Operations Using Inherent Computation Capability of NAND Flash Memory

Bulk bitwise operations, i.e., bitwise operations on large bit vectors, are prevalent in a wide range of important application domains, including databases, graph processing, genome analysis, cryptography, and hyper-dimensional computing. In conventional systems, the performance and energy efficiency of bulk bitwise operations are bottlenecked by data movement between the compute units and the memory hierarchy. In-flash processing (i.e., processing data inside NAND flash chips) has a high potential to accelerate bulk bitwise operations by fundamentally reducing data movement through the entire memory hierarchy. We identify two key limitations of the state-of-the-art in-flash processing technique for bulk bitwise operations; (i) it falls short of maximally exploiting the bit-level parallelism of bulk bitwise operations; (ii) it is unreliable because it does not consider the highly error-prone nature of NAND flash memory. We propose Flash-Cosmos (Flash Computation with One-Shot Multi-Operand Sensing), a new in-flash processing technique that significantly increases the performance and energy efficiency of bulk bitwise operations while providing high reliability. Flash-Cosmos introduces two key mechanisms that can be easily supported in modern NAND flash chips: (i) Multi-Wordline Sensing (MWS), which enables bulk bitwise operations on a large number of operands with a single sensing operation, and (ii) Enhanced SLC-mode Programming (ESP), which enables reliable computation inside NAND flash memory. We demonstrate the feasibility of performing bulk bitwise operations with high reliability in Flash-Cosmos by testing 160 real 3D NAND flash chips. Our evaluation shows that Flash-Cosmos improves average performance and energy efficiency by 3.5x/32x and 3.3x/95x, respectively, over the state-of-the-art in-flash/outside-storage processing techniques across three real-world applications.

cs.AR↗

arXiv2022

FastRemap: A Tool for Quickly Remapping Reads between Genome Assemblies

A genome read data set can be quickly and efficiently remapped from one reference to another similar reference (e.g., between two reference versions or two similar species) using a variety of tools, e.g., the commonly-used CrossMap tool. With the explosion of available genomic data sets and references, high-performance remapping tools will be even more important for keeping up with the computational demands of genome assembly and analysis. We provide FastRemap, a fast and efficient tool for remapping reads between genome assemblies. FastRemap provides up to a 7.82$\times$ speedup (6.47$\times$, on average) and uses as low as 61.7% (80.7%, on average) of the peak memory consumption compared to the state-of-the-art remapping tool, CrossMap. FastRemap is written in C++. The source code and user manual are freely available at: github.com/CMU-SAFARI/FastRemap. Docker image available at: https://hub.docker.com/r/alkanlab/fast. Also available in Bioconda at: https://anaconda.org/bioconda/fastremap-bio.

q-bio.GN↗

arXiv2022

Demeter: A Fast and Energy-Efficient Food Profiler using Hyperdimensional Computing in Memory

Food profiling is an essential step in any food monitoring system needed to prevent health risks and potential frauds in the food industry. Significant improvements in sequencing technologies are pushing food profiling to become the main computational bottleneck. State-of-the-art profilers are unfortunately too costly for food profiling. Our goal is to design a food profiler that solves the main limitations of existing profilers, namely (1) working on massive data structures and (2) incurring considerable data movement for a real-time monitoring system. To this end, we propose Demeter, the first platform-independent framework for food profiling. Demeter overcomes the first limitation through the use of hyperdimensional computing (HDC) and efficiently performs the accurate few-species classification required in food profiling. We overcome the second limitation by using an in-memory hardware accelerator for Demeter (named Acc-Demeter) based on memristor devices. Acc-Demeter actualizes several domain-specific optimizations and exploits the inherent characteristics of memristors to improve the overall performance and energy consumption of Acc-Demeter. We compare Demeter's accuracy with other industrial food profilers using detailed software modeling. We synthesize Acc-Demeter's required hardware using UMC's 65nm library by considering an accurate PCM model based on silicon-based prototypes. Our evaluations demonstrate that Acc-Demeter achieves a (1) throughput improvement of 192x and 724x and (2) memory reduction of 36x and 33x compared to Kraken2 and MetaCache (2 state-of-the-art profilers), respectively, on typical food-related databases. Demeter maintains an acceptable profiling accuracy (within 2% of existing tools) and incurs a very low area overhead.

cs.AR↗

arXiv2022

Machine Learning Training on a Real Processing-in-Memory System

Training machine learning algorithms is a computationally intensive process, which is frequently memory-bound due to repeatedly accessing large training datasets. As a result, processor-centric systems (e.g., CPU, GPU) suffer from costly data movement between memory units and processing units, which consumes large amounts of energy and execution cycles. Memory-centric computing systems, i.e., computing systems with processing-in-memory (PIM) capabilities, can alleviate this data movement bottleneck. Our goal is to understand the potential of modern general-purpose PIM architectures to accelerate machine learning training. To do so, we (1) implement several representative classic machine learning algorithms (namely, linear regression, logistic regression, decision tree, K-means clustering) on a real-world general-purpose PIM architecture, (2) characterize them in terms of accuracy, performance and scaling, and (3) compare to their counterpart implementations on CPU and GPU. Our experimental evaluation on a memory-centric computing system with more than 2500 PIM cores shows that general-purpose PIM architectures can greatly accelerate memory-bound machine learning workloads, when the necessary operations and datatypes are natively supported by PIM hardware. To our knowledge, our work is the first one to evaluate training of machine learning algorithms on a real-world general-purpose PIM architecture.

cs.AR↗