An official website of the United States government
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
BurstZ: a bandwidth-efficient scientific computing accelerator platform for large-scale data
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
- Award ID(s):
- 1908507
- PAR ID:
- 10195756
- Date Published:
- Journal Name:
- Proceedings of the 34th ACM International Conference on Supercomputing
- Page Range / eLocation ID:
- 1 to 12
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)With ever-increasing volumes of scientific floating-point data being produced by high-performance computing applications, significantly reducing scientific floating-point data size is critical, and error-controlled lossy compressors have been developed for years. None of the existing scientific floating-point lossy data compressors, however, support effective fixed-ratio lossy compression. Yet fixed-ratio lossy compression for scientific floating-point data not only compresses to the requested ratio but also respects a user-specified error bound with higher fidelity. In this paper, we present FRaZ: a generic fixed-ratio lossy compression framework respecting user-specified error constraints. The contribution is twofold. (1) We develop an efficient iterative approach to accurately determine the appropriate error settings for different lossy compressors based on target compression ratios. (2) We perform a thorough performancemore » « less
-
With ever-increasing volumes of scientific floating-point data being produced by high-performance computing applications, significantly reducing scientific floating-point data size is critical, and error-controlled lossy compressors have been developed for years. None of the existing scientific floating-point lossy data compressors, however, support effective fixed-ratio lossy compression. Yet fixed-ratio lossy compression for scientific floating-point data not only compresses to the requested ratio but also respects a user-specified error bound with higher fidelity. In this paper, we present FRaZ: a generic fixed-ratio lossy compression framework respecting user-specified error constraints. The contribution is twofold. (1) We develop an efficient iterative approach to accurately determine the appropriate error settings for different lossy compressors based on target compression ratios. (2) We perform a thorough performance and accuracy evaluation for our proposed fixed-ratio compression framework with multiple state-of-the-art error-controlled lossy compressors, using several real-world scientific floating-point datasets from different domains. Experiments show that FRaZ effectively identifies the optimum error setting in the entire error setting space of any given lossy compressor. While fixed-ratio lossy compression is slower than fixed-error compression, it provides an important new lossy compression technique for users of very large scientific floating-point datasets.more » « less
-
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
-
The continued growth in the processing power of FPGAs coupled with high bandwidth memories (HBM), makes systems like the Xilinx U280 credible platforms for linear solvers which often dominate the run time of scientific and engineering applications. In this paper, we present Callipepla, an accelerator for a preconditioned conjugate gradient linear solver (CG). FPGA acceleration of CG faces three challenges: (1) how to support an arbitrary problem and terminate acceleration processing on the fly, (2) how to coordinate long-vector data flow among processing modules, and (3) how to save off-chip memory bandwidth and maintain double (FP64) precision accuracy. To tackle the three challenges, we present (1) a stream-centric instruction set for efficient streaming processing and control, (2) vector streaming reuse (VSR) and decentralized vector flow scheduling to coordinate vector data flow among modules and further reduce off-chip memory access latency with a double memory channel design, and (3) a mixed precision scheme to save bandwidth yet still achieve effective double precision quality solutions. To the best of our knowledge, this is the first work to introduce the concept of VSR for data reusing between on-chip modules to reduce unnecessary off-chip accesses and enable modules working in parallel for FPGA accelerators. We prototype the accelerator on a Xilinx U280 HBM FPGA. Our evaluation shows that compared to the Xilinx HPC product, the XcgSolver, Callipepla achieves a speedup of 3.94×, 3.36× higher throughput, and 2.94× better energy efficiency. Compared to an NVIDIA A100 GPU which has 4× the memory bandwidth of Callipepla, we still achieve 77% of its throughput with 3.34× higher energy efficiency. The code is available at https://github.com/UCLA-VAST/Callipepla.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3392717.3392746
