skip to main content

Title: Predictable Accelerator Design with Time-Sensitive Affine Types
Field-programmable gate arrays (FPGAs) provide an opportunity to co-design applications with hardware accelerators, yet they remain difficult to program. High-level synthesis (HLS) tools promise to raise the level of abstraction by compiling C or C++ to accelerator designs. Repurposing legacy software languages, however, requires complex heuristics to map imperative code onto hardware structures. We find that the black-box heuristics in HLS can be unpredictable: changing parameters in the program that should improve performance can counterintuitively yield slower and larger designs. This paper proposes a type system that restricts HLS to programs that can predictably compile to hardware accelerators. The key idea is to model consumable hardware resources with a time-sensitive affine type system that prevents simultaneous uses of the same hardware structure. We implement the type system in Dahlia, a language that compiles to HLS C++, and show that it can reduce the size of HLS parameter spaces while accepting Pareto-optimal designs.
Authors:
; ; ; ; ; ; ; ;
Award ID(s):
1845952
Publication Date:
NSF-PAR ID:
10148694
Journal Name:
Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation
ISSN:
1531-7102
Sponsoring Org:
National Science Foundation
More Like this
  1. When quantum computers become scalable and reliable, they are likely to break all public-key cryptography standards, such as RSA and Elliptic Curve Cryptography. The projected threat of quantum computers has led the U.S. National Institute of Standards and Technology (NIST) to an effort aimed at replacing existing public-key cryptography standards with new quantum-resistant alternatives. In December 2017, 69 candidates were accepted by NIST to Round 1 of the NIST Post-Quantum Cryptography (PQC) standardization process. NTRUEncrypt is one of the most well-known PQC algorithms that has withstood cryptanalysis. The speed of NTRUEncrypt in software, especially on embedded software platforms, is limited by the long execution time of its primary operation, polynomial multiplication. In this paper, we investigate speeding up NTRUEncrypt using software/hardware codesign on a Xilinx Zynq UltraScale+ multiprocessor system-on-chip (MPSoC). Polynomial multiplication is implemented in the Programmable Logic (PL) of Zynq using two approaches: traditional Register-Transfer Level (RTL) and High-Level Synthesis (HLS). The remaining operations of NTRUEncrypt are executed in software on the Processing System (PS) of Zynq, using the bare-metal mode. The speed-up of our software/hardware codesigns vs. purely software implementations is determined experimentally and analyzed in the paper. The results are reported for the RTL-based and HLS-based hardwaremore »accelerators, and compared to the best available software implementation, included in the NIST submission package. The speed-ups for encryption were 2.4 and 3.9, depending on the selected parameter set. For decryption, the corresponding speed-ups were 4.0 and 6.8. In addition, for the polynomial multiplication operation itself, the speed up was in excess of 75. Our code for the NTRUEncrypt polynomial multiplier accelerator is being made open-source for further evaluation on multiple software/hardware platforms.« less
  2. When quantum computers become scalable and reliable, they are likely to break all public-key cryptography standards, such as RSA and Elliptic Curve Cryptography. The projected threat of quantum computers has led the U.S. National Institute of Standards and Technology (NIST) to an effort aimed at replacing existing public-key cryptography standards with new quantum-resistant alternatives. In December 2017, 69 candidates were accepted by NIST to Round 1 of the NIST Post-Quantum Cryptography (PQC) standardization process. NTRUEncrypt is one of the most well-known PQC algorithms that has withstood cryptanalysis. The speed of NTRUEncrypt in software, especially on embedded software platforms, is limited by the long execution time of its primary operation, polynomial multiplication. In this paper, we investigate speeding up NTRUEncrypt using software/hardware codesign on a Xilinx Zynq UltraScale+ multiprocessor system-on-chip (MPSoC). Polynomial multiplication is implemented in the Programmable Logic (PL) of Zynq using two approaches: traditional Register-Transfer Level (RTL) and High-Level Synthesis (HLS). The remaining operations of NTRUEncrypt are executed in software on the Processing System (PS) of Zynq, using the bare-metal mode. The speed-up of our software/hardware codesigns vs. purely software implementations is determined experimentally and analyzed in the paper. The results are reported for the RTL-based and HLS-based hardwaremore »accelerators, and compared to the best available software implementation, included in the NIST submission package. The speed-ups for encryption were 2.4 and 3.9, depending on the selected parameter set. For decryption, the corresponding speed-ups were 4.0 and 6.8. In addition, for the polynomial multiplication operation itself, the speed up was in excess of 75. Our code for the NTRUEncrypt polynomial multiplier accelerator is being made open-source for further evaluation on multiple software/hardware platforms.« less
  3. As specialized hardware accelerators like FPGAs become a prominent part of the current computing landscape, software applications are increasingly constructed to leverage heterogeneous architectures. Such a trend is already happening in the domain of machine learning and Internet-of-Things (IoT) systems built on edge devices. Yet, debugging and testing methods for heterogeneous applications are currently lacking. These applications may look similar to regular C/C++ code but include hardware synthesis details in terms of preprocessor directives. Therefore, their behavior under heterogeneous architectures may diverge significantly from CPU due to hardware synthesis details. Further, the compilation and hardware simulation cycle takes an enormous amount of time, prohibiting frequent invocations required for fuzz testing. We propose a novel fuzz testing technique, called HeteroFuzz, designed to specifically target heterogeneous applications and to detect platform-dependent divergence. The key essence of HeteroFuzz is that it uses a three-pronged approach to reduce the long latency of repetitively invoking a hardware simulator on a heterogeneous application. First, in addition to monitoring code coverage as a fuzzing guidance mechanism, we analyze synthesis pragmas in kernel code and monitor accelerator-relevant value spectra. Second, we design dynamic probabilistic mutations to increase the chance of hitting divergent behavior under different platforms. Third, wemore »memorize the boundaries of seen kernel inputs and skip HLS simulator invocation if it can expose only redundant divergent behavior. We evaluate HeteroFuzz on seven real-world heterogeneous applications with FPGA kernels. HeteroFuzz is 754X faster in exposing the same set of distinct divergence symptoms than naive fuzzing. Probabilistic mutations contribute to 17.5X speed up than the one without. Selective invocation of HLS simulation contributes to 8.8X speed up than the one without.« less
  4. FPGAs offer a heterogenous compute solution to the continuous de- sire for increased performance by enabling the creation of application- specific hardware that accelerates computation. While the barrier to entry has historically been steep, advances in High Level Synthe- sis (HLS) are making FPGAs more accessible. Specifically, the Intel FPGA OpenCL SDK allows software designers to abstract away low level details of architecting hardware on an FPGA and allows them to author computational kernels in a higher level language. Furthermore, Intel has developed a system that incorporates both a multicore Xeon CPU and Arria 10 FPGA into the same chip package as part of the Heterogeneous Accelerator Research Program (HARP) that can be targeted by their SDK. In this work, we target the second iteration of the HARP platform (HARPv2) using HLS through porting of OpenCL kernels originally written for FPGAs connected via a PCIe bus. We evaluate the HARPv2 system’s performance against previously reported results, explore the portability of kernels through a hardware design space search, and empirically show the benefits of using the shared virtual memory (SVM) abstraction over explicit reads and writes.
  5. 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 towardsmore »democratizing customizable computing by enabling software programmers to design efficient FPGA accelerators.« less