NGEMM: Optimizing GEMM for Deep Learning via Compiler-based Techniques
원문 초록
Quantization has emerged to be an effective way to significantly boost the performance of deep neural networks (DNNs) by utilizing low-bit computations. Despite having lower numerical precision, quantized DNNs are able to reduce both memory bandwidth and computation cycles with little losses of accuracy. Integer GEMM (General Matrix Multiplication) is critical to running quantized DNN models efficiently, as GEMM operations often dominate the computations in these models. Various approaches have been developed by leveraging techniques such as vectorization and memory layout to improve the performance of integer GEMM. However, these existing approaches are not fast enough in certain scenarios. We developed NGEMM, a compiler-based GEMM implementation for accelerating lower-precision training and inference. NGEMM has better use of the vector units by avoiding unnecessary vector computation that is introduced during tree reduction. We compared NGEMM's performance with the state-of-art BLAS libraries such as MKL. Our experimental results showed that NGEMM outperformed MKL non-pack and pack version by an average of 1.86x and 1.16x, respectively. We have applied NGEMM to a number of production services in Microsoft.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group Quantization
Compress Any Segment Anything Model (SAM)
OpenDPDv2: A Unified Learning and Optimization Framework for Neural Network Digital Predistortion
EdgeCodec: Onboard Lightweight High Fidelity Neural Compressor with Residual Vector Quantization
any4: Learned 4-bit Numeric Representation for LLMs
CycleVAR: Repurposing Autoregressive Model for Unsupervised One-Step Image Translation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.