An official website of the United States government
We present BurstZ, a bandwidth-efficient accelerator platform for scientific computing. While accelerators such as GPUs and FPGAs provide enormous computing capabilities, their effectiveness quickly deteriorates once the working set becomes larger than the on-board memory capacity, causing the performance to become bottlenecked either by the communication bandwidth between the host and the accelerator. Compression has not been very useful in solving this issue due to the difficulty of efficiently compressing floating point numbers, which scientific data often consists of. Most compression algorithms are either ineffective with floating point numbers, or has a high performance overhead. BurstZ is an FPGA-based accelerator platform which addresses the bandwidth issue via a novel hardware-optimized floating point compression algorithm, which we call sZFP. We demonstrate that BurstZ can completely remove the communication bottleneck for accelerators, using a 3D stencil-code accelerator implemented on a prototype BurstZ implementation. Evaluated against hand-optimized implementations of stencil code accelerators of the same architecture, our BurstZ prototype outperformed an accelerator without compression by almost 4X, and even an accelerator with enough memory for the entire dataset by over 2X. BurstZ improved communication efficiency so much, our prototype was even able to outperform the upper limit projected performance of an optimized stencil core with ideal memory access characteristics, by over 2X.
more »
« less
AutoDSE: Enabling Software Programmers to Design Efficient FPGA Accelerators
Adopting FPGA as an accelerator in datacenters is becoming mainstream for customized computing, but the fact that FPGAs are hard to program creates a steep learning curve for software programmers. Even with the help of high-level synthesis (HLS) , accelerator designers still have to manually perform code reconstruction and cumbersome parameter tuning to achieve optimal performance. While many learning models have been leveraged by existing work to automate the design of efficient accelerators, the unpredictability of modern HLS tools becomes a major obstacle for them to maintain high accuracy. To address this problem, we propose an automated DSE framework— AutoDSE —that leverages a bottleneck-guided coordinate optimizer to systematically find a better design point. AutoDSE detects the bottleneck of the design in each step and focuses on high-impact parameters to overcome it. The experimental results show that AutoDSE is able to identify the design point that achieves, on the geometric mean, 19.9× speedup over one CPU core for MachSuite and Rodinia benchmarks. Compared to the manually optimized HLS vision kernels in Xilinx Vitis libraries, AutoDSE can reduce their optimization pragmas by 26.38× while achieving similar performance. With less than one optimization pragma per design on average, we are making progress towards democratizing customizable computing by enabling software programmers to design efficient FPGA accelerators.
more »
« less
- PAR ID:
- 10356794
- Date Published:
- Journal Name:
- ACM Transactions on Design Automation of Electronic Systems
- Volume:
- 27
- Issue:
- 4
- ISSN:
- 1084-4309
- Page Range / eLocation ID:
- 1 to 27
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Deep Neural Networks (DNNs) have been successfully applied in many fields. Considering performance, flexibility, and energy efficiency, Field Programmable Gate Array (FPGA) based accelerator for DNNs is a promising solution. The existing frameworks however lack the possibility of reusability and friendliness to design a new network with minimum efforts. Modern high-level synthesis (HLS) tools greatly reduce the turnaround time of designing and implementing complex FPGA-based accelerators. This paper presents a framework for hardware accelerator for DNNs using high level specification. A novel architecture is introduced that maximizes data reuse and external memory bandwidth. This framework allows to generate a scalable HLS code for a given pre-trained model that can be mapped to different FPGA platforms. Various HLS compiler optimizations have been applied to the code to produce efficient implementation and high resource utilization. The framework achieves a peak performance of 23 frames per second for SqueezeNet on Xilinx Alveo u250 board.more » « less
-
Deep-Learning has become a dominant computing paradigm across a broad range of application domains. Different architectures of Deep-Networks like CNN, MLP, and RNN have emerged as the prominent machine-learning approaches for today’s application domains. These architectures are heavily data-dependent, requiring frequent access to memory. As a result, these applications suffer the most from the memory bottleneck of the von Neumann architectures. There is an imminent need for memory-centric architectures for deep-learning and big-data analytic applications that are memory intensive. Modern Field Programmable Gate Arrays (FPGAs) are ideal programmable substrates for creating customized Processor in/near Memory (PIM) accelerators. Modern FPGAs contain 100s of Mbits of dual-ported SRAM in the form of disaggregated, configurable Block RAMs (BRAMs). These BRAMs contain TB/s of available internal bandwidth. Unfortunately, developing FPGA-based accelerators for deep learning is not a simple task and demands the utilization of specialized tools provided by the FPGA vendors. It requires expertise in low-level hardware microarchitecture design. These are often not available to most researchers in the field of deep learning. Even with the ongoing improvements in High-Level Synthesis (HLS) tools, the requirement for hardware-specific design knowledge cannot be completely eliminated. This research developed a new reconfigurable memory-centric architecture and design approach that opens the advantages of FPGAs and Processor-in-Memory architecture to memory-intensive applications. Due to its high-performance and scalable memory-centric design, this architecture can deliver the highest speed and the lowest latency achievable from an FPGA overcoming the memory bottleneck.more » « less
-
Real-time systems are widely applied in different areas like autonomous vehicles, where safety is the key metric. However, on the FPGA platform, most of the prior accelerator frameworks omit discussing the schedulability in such real-time safety-critical systems, leaving deadlines unmet, which can lead to catastrophic system failures. To address this, we propose the ART framework, a hardware-software co-design approach that transforms baseline accelerators into “real-time guaranteed" accelerators. On the software side, ART performs schedulability analysis and preemption point placement, optimizing task scheduling to meet deadlines and enhance throughput. On the hardware side, ART integrates the Global Earliest Deadline First (GEDF) scheduling algorithm, implements preemption, and conducts source code transformation to transform baseline HLS-based accelerators into designs targeted for real-time systems capable of saving and resuming tasks. ART also includes integration, debugging, and testing tools for full-system implementation. We demonstrate the methodology of ART on two kinds of popular accelerator models and evaluate on AMD Versal VCK190 platform, where ART meets schedulability requirements that baseline accelerators fail. ART is lightweight, utilizing <0.5% resources. With about 100 lines of user input, ART generates about 2.5k lines of accelerator code, making it a push-button solution.more » « less
-
Dense matrix multiply (MM) serves as one of the most heavily used kernels in deep learning applications. To cope with the high computation demands of these applications, heterogeneous architectures featuring both FPGA and dedicated ASIC accelerators have emerged as promising platforms. For example, the AMD/Xilinx Versal ACAP architecture combines general-purpose CPU cores and programmable logic (PL) with AI Engine processors (AIE) optimized for AI/ML. An array of 400 AI Engine processors executing at 1 GHz can theoretically provide up to 6.4 TFLOPs performance for 32-bit floating-point (fp32) data. However, machine learning models often contain both large and small MM operations. While large MM operations can be parallelized efficiently across many cores, small MM operations typically cannot. In our investigation, we observe that executing some small MM layers from the BERT natural language processing model on a large, monolithic MM accelerator in Versal ACAP achieved less than 5% of the theoretical peak performance. Therefore, one key question arises: How can we design accelerators to fully use the abundant computation resources under limited communication bandwidth for end-to-end applications with multiple MM layers of diverse sizes? We identify the biggest system throughput bottleneck resulting from the mismatch of massive computation resources of one monolithic accelerator and the various MM layers of small sizes in the application. To resolve this problem, we propose the CHARM framework to compose multiple diverse MM accelerator architectures working concurrently towards different layers within one application. CHARM includes analytical models which guide design space exploration to determine accelerator partitions and layer scheduling. To facilitate the system designs, CHARM automatically generates code, enabling thorough onboard design verification. We deploy the CHARM framework for four different deep learning applications, including BERT, ViT, NCF, MLP, on the AMD/Xilinx Versal ACAP VCK190 evaluation board. Our experiments show that we achieve 1.46 TFLOPs, 1.61 TFLOPs, 1.74 TFLOPs, and 2.94 TFLOPs inference throughput for BERT, ViT, NCF, MLP, respectively, which obtain 5.40x, 32.51x, 1.00x and 1.00x throughput gains compared to one monolithic accelerator.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1145/3494534
