Search

Deep neural networks (DNNs) are of critical use in different domains. To accelerate DNN computation, tensor compilers are proposed to generate efficient code on different domain-specific accelerators. Existing tensor compilers mainly focus on optimizing computation efficiency. However, memory access is becoming a key performance bottleneck because the computational performance of accelerators is increasing much faster than memory performance. The lack of direct description of memory access and data dependence in current tensor compilers' intermediate representation (IR) brings significant challenges to generate memory-efficient code. In this paper, we propose IntelliGen, a tensor compiler that can generate high-performance code for memory-intensive operators by considering both computation and data movement optimizations. IntelliGen represent a DNN program using GIR, which includes primitives indicating its computation, data movement, and parallel strategies. This information will be further composed as an instruction-level dataflow graph to perform holistic optimizations by searching different memory access patterns and computation operations, and generating memory-efficient code on different hardware. We evaluate IntelliGen on NVIDIA GPU, AMD GPU, and Cambricon MLU, showing speedup up to 1.97x, 2.93x, and 16.91x(1.28x, 1.23x, and 2.31x on average), respectively, compared to current most performant frameworks., Comment: 12 pages, 14 figures

Boosting the runtime performance of deep neural networks (DNNs) is critical due to their wide adoption in real-world tasks. Existing approaches to optimizing the tensor algebra expression of a DNN only consider expressions representable by a fixed set of predefined operators, missing possible optimization opportunities between general expressions. We propose OLLIE, the first derivation-based tensor program optimizer. OLLIE optimizes tensor programs by leveraging transformations between general tensor algebra expressions, enabling a significantly larger expression search space that includes those supported by prior work as special cases. OLLIE uses a hybrid derivation-based optimizer that effectively combines explorative and guided derivations to quickly discover highly optimized expressions. Evaluation on seven DNNs shows that OLLIE can outperform existing optimizers by up to 2.73$\times$ (1.46$\times$ on average) on an A100 GPU and up to 2.68$\times$ (1.51$\times$) on a V100 GPU, respectively.

We carried out a comparative study on laser cutting of fiber-reinforced plastic and its honeycomb sandwich structure widely used in aeronautic industry. The influence of pulse duration (CW, ns, ps and fs), wavelength (MIR, NIR, and NUV) on cutting quality (characterized by the HAZ) and productivity was experimentally studied. It was demonstrated that for either the aramid fiber reinforced plastics/polymers (AFRPs) or high-modulus carbon fiber reinforced plastics/polymers (CFRPs) and its honeycomb sandwich structure, tradition laser cutting was unable to satisfy the requirements of the appearance qualities in the astronautic industry due to the thermal damage induced discoloration, in spite of the less extension of damage than the contact machining used currently. The pulse duration effects on HAZ, the wavelength effects on cutting quality and productivity, the differences and similarities for low-HAZ cutting of AFRPs and high-modulus CFRPs, was studied, and the possible physical origins was discussed.

Morphological evolution of ripple on nickel surface induced by temporally shaped femtosecond laser irradiation were studied and compared with titanium. It was revealed that the transformation of single-pulse irradiation into double-pulse irradiation can exert very different influence on ripple morphology evolution for different metals. For nickel, the double-pulse irradiation resulted in the growth in rippled area, ripple period and ripple contrast, compared with the single-pulse irradiation, while double-pulse processing of titanium leads to reductions in in rippled area and ripple period. The contrasted influence of Te on electron-phonon coupling factor (G) for the two metals was the primary factor for the different behaviors of ripple morphology evolution with temporal pulse shaping.