arXiv subjects

David Kaeli

Publications and source records attributed to David Kaeli.

At least 37 records · Page 2Linked to original sources

arXiv2024

NeuraChip: Accelerating GNN Computations with a Hash-based Decoupled Spatial Accelerator

Graph Neural Networks (GNNs) are emerging as a formidable tool for processing non-euclidean data across various domains, ranging from social network analysis to bioinformatics. Despite their effectiveness, their adoption has not been pervasive because of scalability challenges associated with large-scale graph datasets, particularly when leveraging message passing. To tackle these challenges, we introduce NeuraChip, a novel GNN spatial accelerator based on Gustavson's algorithm. NeuraChip decouples the multiplication and addition computations in sparse matrix multiplication. This separation allows for independent exploitation of their unique data dependencies, facilitating efficient resource allocation. We introduce a rolling eviction strategy to mitigate data idling in on-chip memory as well as address the prevalent issue of memory bloat in sparse graph computations. Furthermore, the compute resource load balancing is achieved through a dynamic reseeding hash-based mapping, ensuring uniform utilization of computing resources agnostic of sparsity patterns. Finally, we present NeuraSim, an open-source, cycle-accurate, multi-threaded, modular simulator for comprehensive performance analysis. Overall, NeuraChip presents a significant improvement, yielding an average speedup of 22.1x over Intel's MKL, 17.1x over NVIDIA's cuSPARSE, 16.7x over AMD's hipSPARSE, and 1.5x over prior state-of-the-art SpGEMM accelerator and 1.3x over GNN accelerator. The source code for our open-sourced simulator and performance visualizer is publicly accessible on GitHub https://neurachip.us

cs.AR↗

arXiv2024

MaxK-GNN: Extremely Fast GPU Kernel Design for Accelerating Graph Neural Networks Training

In the acceleration of deep neural network training, the GPU has become the mainstream platform. GPUs face substantial challenges on GNNs, such as workload imbalance and memory access irregularities, leading to underutilized hardware. Existing solutions such as PyG, DGL with cuSPARSE, and GNNAdvisor frameworks partially address these challenges but memory traffic is still significant. We argue that drastic performance improvements can only be achieved by the vertical optimization of algorithm and system innovations, rather than treating the speedup optimization as an "after-thought" (i.e., (i) given a GNN algorithm, designing an accelerator, or (ii) given hardware, mainly optimizing the GNN algorithm). In this paper, we present MaxK-GNN, an advanced high-performance GPU training system integrating algorithm and system innovation. (i) We introduce the MaxK nonlinearity and provide a theoretical analysis of MaxK nonlinearity as a universal approximator, and present the Compressed Balanced Sparse Row (CBSR) format, designed to store the data and index of the feature matrix after nonlinearity; (ii) We design a coalescing enhanced forward computation with row-wise product-based SpGEMM Kernel using CBSR for input feature matrix fetching and strategic placement of a sparse output accumulation buffer in shared memory; (iii) We develop an optimized backward computation with outer product-based and SSpMM Kernel. We conduct extensive evaluations of MaxK-GNN and report the end-to-end system run-time. Experiments show that MaxK-GNN system could approach the theoretical speedup limit according to Amdahl's law. We achieve comparable accuracy to SOTA GNNs, but at a significantly increased speed: 3.22/4.24 times speedup (vs. theoretical limits, 5.52/7.27 times) on Reddit compared to DGL and GNNAdvisor implementations.

cs.LG↗

arXiv2024

Data Transfer Optimizations for Host-CPU and Accelerators in AXI4MLIR

As custom hardware accelerators become more prevalent, it becomes increasingly important to automatically generate efficient host-driver code that can fully leverage the capabilities of these accelerators. This approach saves time and reduces the likelihood of errors that can occur during manual implementation. AXI4MLIR extends the MLIR compiler framework to generate host-driver code for custom accelerators for linear algebra problems. By leveraging specific compiler optimizations, we can further increase accelerator utilization. In this work we offer two key observations through a MatMul accelerator case study. First, the accelerator's compute core utilization is less than 10%, and second, the critical latency bottleneck is caused by copying data between the heap and memory-mapped DMA buffers. We identify a set of missing host code optimizations to improve the under-utilization and the latency bottleneck. Therefore, we propose three key host-code data-movement-related optimizations, extending AXI4MLIR. The optimizations provide DMA-based data allocation, coalescing of DMA transfers, and pipelining of the accelerator's load, compute, and store stages.

cs.PL↗

arXiv2023

AXI4MLIR: User-Driven Automatic Host Code Generation for Custom AXI-Based Accelerators

This paper addresses the need for automatic and efficient generation of host driver code for arbitrary custom AXI-based accelerators targeting linear algebra algorithms, an important workload in various applications, including machine learning and scientific computing. While existing tools have focused on automating accelerator prototyping, little attention has been paid to the host-accelerator interaction. This paper introduces AXI4MLIR, an extension of the MLIR compiler framework designed to facilitate the automated generation of host-accelerator driver code. With new MLIR attributes and transformations, AXI4MLIR empowers users to specify accelerator features (including their instructions) and communication patterns and exploit the host memory hierarchy. We demonstrate AXI4MLIR's versatility across different types of accelerators and problems, showcasing significant CPU cache reference reductions (up to 56%) and up to a 1.65x speedup compared to manually optimized driver code implementations. AXI4MLIR implementation is open-source and available at: https://github.com/AXI4MLIR/axi4mlir.

cs.PL↗

arXiv2023

Memory Efficient Multithreaded Incremental Segmented Sieve Algorithm

Prime numbers are fundamental in number theory and play a significant role in various areas, from pure mathematics to practical applications, including cryptography. In this contribution, we introduce a multithreaded implementation of the Segmented Sieve algorithm. In our implementation, instead of handling large prime ranges in one iteration, the sieving process is broken down incrementally, which theoretically eliminates the challenges of working with large numbers, and can reduce memory usage, providing overall more efficient multi-core utilization over extended computations.

cs.PF↗

arXiv2023

GME: GPU-based Microarchitectural Extensions to Accelerate Homomorphic Encryption

Fully Homomorphic Encryption (FHE) enables the processing of encrypted data without decrypting it. FHE has garnered significant attention over the past decade as it supports secure outsourcing of data processing to remote cloud services. Despite its promise of strong data privacy and security guarantees, FHE introduces a slowdown of up to five orders of magnitude as compared to the same computation using plaintext data. This overhead is presently a major barrier to the commercial adoption of FHE. In this work, we leverage GPUs to accelerate FHE, capitalizing on a well-established GPU ecosystem available in the cloud. We propose GME, which combines three key microarchitectural extensions along with a compile-time optimization to the current AMD CDNA GPU architecture. First, GME integrates a lightweight on-chip compute unit (CU)-side hierarchical interconnect to retain ciphertext in cache across FHE kernels, thus eliminating redundant memory transactions. Second, to tackle compute bottlenecks, GME introduces special MOD-units that provide native custom hardware support for modular reduction operations, one of the most commonly executed sets of operations in FHE. Third, by integrating the MOD-unit with our novel pipelined $64$-bit integer arithmetic cores (WMAC-units), GME further accelerates FHE workloads by $19\%$. Finally, we propose a Locality-Aware Block Scheduler (LABS) that exploits the temporal locality available in FHE primitive blocks. Incorporating these microarchitectural features and compiler optimizations, we create a synergistic approach achieving average speedups of $796\times$, $14.2\times$, and $2.3\times$ over Intel Xeon CPU, NVIDIA V100 GPU, and Xilinx FPGA implementations, respectively.

cs.CR↗

arXiv2023

Achieving on-Mobile Real-Time Super-Resolution with Neural Architecture and Pruning Search

Though recent years have witnessed remarkable progress in single image super-resolution (SISR) tasks with the prosperous development of deep neural networks (DNNs), the deep learning methods are confronted with the computation and memory consumption issues in practice, especially for resource-limited platforms such as mobile devices. To overcome the challenge and facilitate the real-time deployment of SISR tasks on mobile, we combine neural architecture search with pruning search and propose an automatic search framework that derives sparse super-resolution (SR) models with high image quality while satisfying the real-time inference requirement. To decrease the search cost, we leverage the weight sharing strategy by introducing a supernet and decouple the search problem into three stages, including supernet construction, compiler-aware architecture and pruning search, and compiler-aware pruning ratio search. With the proposed framework, we are the first to achieve real-time SR inference (with only tens of milliseconds per frame) for implementing 720p resolution with competitive image quality (in terms of PSNR and SSIM) on mobile platforms (Samsung Galaxy S20).

eess.IV↗

arXiv2023

Thought Bubbles: A Proxy into Players' Mental Model Development

Studying mental models has recently received more attention, aiming to understand the cognitive aspects of human-computer interaction. However, there is not enough research on the elicitation of mental models in complex dynamic systems. We present Thought Bubbles as an approach for eliciting mental models and an avenue for understanding players' mental model development in interactive virtual environments. We demonstrate the use of Thought Bubbles in two experimental studies involving 250 participants playing a supply chain game. In our analyses, we rely on Situation Awareness (SA) levels, including perception, comprehension, and projection, and show how experimental manipulations such as disruptions and information sharing shape players' mental models and drive their decisions depending on their behavioral profile. Our results provide evidence for the use of thought bubbles in uncovering cognitive aspects of behavior by indicating how disruption location and availability of information affect people's mental model development and influence their decisions.

cs.HC↗

arXiv2022

Accelerating Polynomial Multiplication for Homomorphic Encryption on GPUs

Homomorphic Encryption (HE) enables users to securely outsource both the storage and computation of sensitive data to untrusted servers. Not only does HE offer an attractive solution for security in cloud systems, but lattice-based HE systems are also believed to be resistant to attacks by quantum computers. However, current HE implementations suffer from prohibitively high latency. For lattice-based HE to become viable for real-world systems, it is necessary for the key bottlenecks - particularly polynomial multiplication - to be highly efficient. In this paper, we present a characterization of GPU-based implementations of polynomial multiplication. We begin with a survey of modular reduction techniques and analyze several variants of the widely-used Barrett modular reduction algorithm. We then propose a modular reduction variant optimized for 64-bit integer words on the GPU, obtaining a 1.8x speedup over the existing comparable implementations. Next, we explore the following GPU-specific improvements for polynomial multiplication targeted at optimizing latency and throughput: 1) We present a 2D mixed-radix, multi-block implementation of NTT that results in a 1.85x average speedup over the previous state-of-the-art. 2) We explore shared memory optimizations aimed at reducing redundant memory accesses, further improving speedups by 1.2x. 3) Finally, we fuse the Hadamard product with neighboring stages of the NTT, reducing the twiddle factor memory footprint by 50%. By combining our NTT optimizations, we achieve an overall speedup of 123.13x and 2.37x over the previous state-of-the-art CPU and GPU implementations of NTT kernels, respectively.

cs.CR↗

arXiv2022

To Trust or to Stockpile: Modeling Human-Simulation Interaction in Supply Chain Shortages

Understanding decision-making in dynamic and complex settings is a challenge yet essential for preventing, mitigating, and responding to adverse events (e.g., disasters, financial crises). Simulation games have shown promise to advance our understanding of decision-making in such settings. However, an open question remains on how we extract useful information from these games. We contribute an approach to model human-simulation interaction by leveraging existing methods to characterize: (1) system states of dynamic simulation environments (with Principal Component Analysis), (2) behavioral responses from human interaction with simulation (with Hidden Markov Models), and (3) behavioral responses across system states (with Sequence Analysis). We demonstrate this approach with our game simulating drug shortages in a supply chain context. Results from our experimental study with 135 participants show different player types (hoarders, reactors, followers), how behavior changes in different system states, and how sharing information impacts behavior. We discuss how our findings challenge existing literature.

cs.HC↗

arXiv2021

SECDA: Efficient Hardware/Software Co-Design of FPGA-based DNN Accelerators for Edge Inference

Edge computing devices inherently face tight resource constraints, which is especially apparent when deploying Deep Neural Networks (DNN) with high memory and compute demands. FPGAs are commonly available in edge devices. Since these reconfigurable circuits can achieve higher throughput and lower power consumption than general purpose processors, they are especially well-suited for DNN acceleration. However, existing solutions for designing FPGA-based DNN accelerators for edge devices come with high development overheads, given the cost of repeated FPGA synthesis passes, reimplementation in a Hardware Description Language (HDL) of the simulated design, and accelerator system integration. In this paper we propose SECDA, a new hardware/software co-design methodology to reduce design time of optimized DNN inference accelerators on edge devices with FPGAs. SECDA combines cost-effective SystemC simulation with hardware execution, streamlining design space exploration and the development process via reduced design evaluation time. As a case study, we use SECDA to efficiently develop two different DNN accelerator designs on a PYNQ-Z1 board, a platform that includes an edge FPGA. We quickly and iteratively explore the system's hardware/software stack, while identifying and mitigating performance bottlenecks. We evaluate the two accelerator designs with four common DNN models, achieving an average performance speedup across models of up to 3.5$\times$ with a 2.9$\times$ reduction in energy consumption over CPU-only inference. Our code is available at https://github.com/gicLAB/SECDA

cs.AR↗

arXiv2021

Daisen: A Framework for Visualizing Detailed GPU Execution

Graphics Processing Units (GPUs) have been widely used to accelerate artificial intelligence, physics simulation, medical imaging, and information visualization applications. To improve GPU performance, GPU hardware designers need to identify performance issues by inspecting a huge amount of simulator-generated traces. Visualizing the execution traces can reduce the cognitive burden of users and facilitate making sense of behaviors of GPU hardware components. In this paper, we first formalize the process of GPU performance analysis and characterize the design requirements of visualizing execution traces based on a survey study and interviews with GPU hardware designers. We contribute data and task abstraction for GPU performance analysis. Based on our task analysis, we propose Daisen, a framework that supports data collection from GPU simulators and provides visualization of the simulator-generated GPU execution traces. Daisen features a data abstraction and trace format that can record simulator-generated GPU execution traces. Daisen also includes a web-based visualization tool that helps GPU hardware designers examine GPU execution traces, identify performance bottlenecks, and verify performance improvement. Our qualitative evaluation with GPU hardware designers demonstrates that the design of Daisen reflects the typical workflow of GPU hardware designers. Using Daisen, participants were able to effectively identify potential performance bottlenecks and opportunities for performance improvement. The open-sourced implementation of Daisen can be found at gitlab.com/akita/vis. Supplemental materials including a demo video, survey questions, evaluation study guide, and post-study evaluation survey are available at osf.io/j5ghq.

cs.DC↗

arXiv2020

Using Undersampling with Ensemble Learning to Identify Factors Contributing to Preterm Birth

In this paper, we propose Ensemble Learning models to identify factors contributing to preterm birth. Our work leverages a rich dataset collected by a NIEHS P42 Center that is trying to identify the dominant factors responsible for the high rate of premature births in northern Puerto Rico. We investigate analytical models addressing two major challenges present in the dataset: 1) the significant amount of incomplete data in the dataset, and 2) class imbalance in the dataset. First, we leverage and compare two types of missing data imputation methods: 1) mean-based and 2) similarity-based, increasing the completeness of this dataset. Second, we propose a feature selection and evaluation model based on using undersampling with Ensemble Learning to address class imbalance present in the dataset. We leverage and compare multiple Ensemble Feature selection methods, including Complete Linear Aggregation (CLA), Weighted Mean Aggregation (WMA), Feature Occurrence Frequency (OFA), and Classification Accuracy Based Aggregation (CAA). To further address missing data present in each feature, we propose two novel methods: 1) Missing Data Rate and Accuracy Based Aggregation (MAA), and 2) Entropy and Accuracy Based Aggregation (EAA). Both proposed models balance the degree of data variance introduced by the missing data handling during the feature selection process while maintaining model performance. Our results show a 42\% improvement in sensitivity versus fallout over previous state-of-the-art methods.

stat.AP↗

arXiv2020

MGPU-TSM: A Multi-GPU System with Truly Shared Memory

The sizes of GPU applications are rapidly growing. They are exhausting the compute and memory resources of a single GPU, and are demanding the move to multiple GPUs. However, the performance of these applications scales sub-linearly with GPU count because of the overhead of data movement across multiple GPUs. Moreover, a lack of hardware support for coherency exacerbates the problem because a programmer must either replicate the data across GPUs or fetch the remote data using high-overhead off-chip links. To address these problems, we propose a multi-GPU system with truly shared memory (MGPU-TSM), where the main memory is physically shared across all the GPUs. We eliminate remote accesses and avoid data replication using an MGPU-TSM system, which simplifies the memory hierarchy. Our preliminary analysis shows that MGPU-TSM with 4 GPUs performs, on average, 3.9x? better than the current best performing multi-GPU configuration for standard application benchmarks.

cs.AR↗

arXiv2020

Hardware/Software Obfuscation against Timing Side-channel Attack on a GPU

GPUs are increasingly being used in security applications, especially for accelerating encryption/decryption. While GPUs are an attractive platform in terms of performance, the security of these devices raises a number of concerns. One vulnerability is the data-dependent timing information, which can be exploited by adversary to recover the encryption key. Memory system features are frequently exploited since they create detectable timing variations. In this paper, our attack model is a coalescing attack, which leverages a critical GPU microarchitectural feature -- the coalescing unit. As multiple concurrent GPU memory requests can refer to the same cache block, the coalescing unit collapses them into a single memory transaction. The access time of an encryption kernel is dependent on the number of transactions. Correlation between a guessed key value and the associated timing samples can be exploited to recover the secret key. In this paper, a series of hardware/software countermeasures are proposed to obfuscate the memory timing side channel, making the GPU more resilient without impacting performance. Our hardware-based approach attempts to randomize the width of the coalescing unit to lower the signal-to-noise ratio. We present a hierarchical Miss Status Holding Register (MSHR) design that can merge transactions across different warps. This feature boosts performance, while, at the same time, secures the execution. We also present a software-based approach to permute the organization of critical data structures, significantly changing the coalescing behavior and introducing a high degree of randomness. Equipped with our new protections, the effort to launch a successful attack is increased up to 1433X . 178X, while also improving encryption/decryption performance up to 7%.

cs.CR↗

arXiv2020

Summarizing CPU and GPU Design Trends with Product Data

Moore's Law and Dennard Scaling have guided the semiconductor industry for the past few decades. Recently, both laws have faced validity challenges as transistor sizes approach the practical limits of physics. We are interested in testing the validity of these laws and reflect on the reasons responsible. In this work, we collect data of more than 4000 publicly-available CPU and GPU products. We find that transistor scaling remains critical in keeping the laws valid. However, architectural solutions have become increasingly important and will play a larger role in the future. We observe that GPUs consistently deliver higher performance than CPUs. GPU performance continues to rise because of increases in GPU frequency, improvements in the thermal design power (TDP), and growth in die size. But we also see the ratio of GPU to CPU performance moving closer to parity, thanks to new SIMD extensions on CPUs and increased CPU core counts.

cs.DC↗

arXiv2020

HALCONE : A Hardware-Level Timestamp-based Cache Coherence Scheme for Multi-GPU systems

While multi-GPU (MGPU) systems are extremely popular for compute-intensive workloads, several inefficiencies in the memory hierarchy and data movement result in a waste of GPU resources and difficulties in programming MGPU systems. First, due to the lack of hardware-level coherence, the MGPU programming model requires the programmer to replicate and repeatedly transfer data between the GPUs' memory. This leads to inefficient use of precious GPU memory. Second, to maintain coherency across an MGPU system, transferring data using low-bandwidth and high-latency off-chip links leads to degradation in system performance. Third, since the programmer needs to manually maintain data coherence, the programming of an MGPU system to maximize its throughput is extremely challenging. To address the above issues, we propose a novel lightweight timestamp-based coherence protocol, HALCONE, for MGPU systems and modify the memory hierarchy of the GPUs to support physically shared memory. HALCONE replaces the Compute Unit (CU) level logical time counters with cache level logical time counters to reduce coherence traffic. Furthermore, HALCONE introduces a novel timestamp storage unit (TSU) with no additional performance overhead in the main memory to perform coherence actions. Our proposed HALCONE protocol maintains the data coherence in the memory hierarchy of the MGPU with minimal performance overhead (less than 1\%). Using a set of standard MGPU benchmarks, we observe that a 4-GPU MGPU system with shared memory and HALCONE performs, on average, 4.6$\times$ and 3$\times$ better than a 4-GPU MGPU system with existing RDMA and with the recently proposed HMG coherence protocol, respectively. We demonstrate the scalability of HALCONE using different GPU counts (2, 4, 8, and 16) and different CU counts (32, 48, and 64 CUs per GPU) for 11 standard benchmarks.

cs.AR↗

arXiv2020

A Smart Background Scheduler for Storage Systems

In today's enterprise storage systems, supported data services such as snapshot delete or drive rebuild can cause tremendous performance interference if executed inline along with heavy foreground IO, often leading to missing SLOs (Service Level Objectives). Typical storage system applications such as web or VDI (Virtual Desktop Infrastructure) follow a repetitive high/low workload pattern that can be learned and forecasted. We propose a priority-based background scheduler that learns this repetitive pattern and allows storage systems to maintain peak performance and in turn meet service level objectives (SLOs) while supporting a number of data services. When foreground IO demand intensifies, system resources are dedicated to service foreground IO requests and any background processing that can be deferred are recorded to be processed in future idle cycles as long as forecast shows that storage pool has remaining capacity. The smart background scheduler adopts a resource partitioning model that allows both foreground and background IO to execute together as long as foreground IOs are not impacted where the scheduler harness any free cycle to clear background debt. Using traces from VDI application, we show how our technique surpasses a method that statically limit the deferred background debt and improve SLO violations from 54.6% when using a fixed background debt watermark to merely a 6.2% if dynamically set by our smart background scheduler.

cs.DC↗