Liancheng Jia
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Liancheng Jia has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Yun Liang
School of Integrated Circuits, Peking University, Beijing, China
4 shared publications - Liqiang Lu
Zhejiang University, Hangzhou, China
3 shared publications - Guangyu Sun
Peking University, Beijing, Beijing, CHINA
1 shared publication - Runsheng Wang
Peking University
1 shared publication - Shengen Yan
SenseTime Research, Beijing, China
1 shared publication - Siyuan Chen
Peking University, China
1 shared publication
Latest Publications
- Open Access
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1271-1288
TileFlow: A Framework for Modeling Fusion Dataflow via Tree-based Analysis
With the increasing size of DNN models and the growing discrepancy between compute performance and memory bandwidth, fusing multiple layers together to reduce off-chip memory access has become a popular approach in dataflow design. However, designing such dataflows requires flexible and accurate performance models to facilitate evaluation, architecture analysis, and design space exploration. Unfortunately, current state-of-the-art performance models are limited to the dataflows of single operator acceleration, making them inapplicable to operator fusion dataflows. In this paper, we propose a framework called TileFlow that models dataflows for operator fusion. We first characterize the design space of fusion dataflows as a 3D space encompassing compute ordering, resource binding, and loop tiling. We then introduce a tile-centric notation to express dataflow designs within this space. Inspired by the tiling structure of fusion dataflows, we present a tree-based approach to analyze two critical performance metrics: data movement volume within the accelerator memory hierarchy and accelerator compute/memory resource usage. Finally, we leverage these metrics to calculate latency and energy consumption. Our evaluation validates TileFlow’s modeling accuracy against both real hardware and state-of-the-art performance models. We use TileFlow to aid in fusion dataflow design and analysis, and it helps us discover fusion dataflows that achieve an average runtime speedup of 1.85 × for self-attention and 1.28 × for convolution chains compared to the state-of-the-art dataflow.
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1271-1288
Open AccessTileFlow: A Framework for Modeling Fusion Dataflow via Tree-based Analysis
With the increasing size of DNN models and the growing discrepancy between compute performance and memory bandwidth, fusing multiple layers together to reduce off-chip memory access has become a popular approach in dataflow design. However, designing such dataflows requires flexible and accurate performance models to facilitate evaluation, architecture analysis, and design space exploration. Unfortunately, current state-of-the-art performance models are limited to the dataflows of single operator acceleration, making them inapplicable to operator fusion dataflows. In this paper, we propose a framework called TileFlow that models dataflows for operator fusion. We first characterize the design space of fusion dataflows as a 3D space encompassing compute ordering, resource binding, and loop tiling. We then introduce a tile-centric notation to express dataflow designs within this space. Inspired by the tiling structure of fusion dataflows, we present a tree-based approach to analyze two critical performance metrics: data movement volume within the accelerator memory hierarchy and accelerator compute/memory resource usage. Finally, we leverage these metrics to calculate latency and energy consumption. Our evaluation validates TileFlow’s modeling accuracy against both real hardware and state-of-the-art performance models. We use TileFlow to aid in fusion dataflow design and analysis, and it helps us discover fusion dataflows that achieve an average runtime speedup of 1.85 × for self-attention and 1.28 × for convolution chains compared to the state-of-the-art dataflow.
- Open Access
- 27 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (6) , 1898-1911
Automatic Generation of Spatial Accelerator for Tensor Algebra
Tensor algebra finds applications in various domains including machine learning applications, data analytics and others. Spatial hardware accelerators are widely used to boost the performance of tensor algebra applications. It has a complex hardware architecture and rich design space. Prior approaches based on manual implementation lead to low programming productivity, making it hard to explore the large design space. In this paper, we propose Tensorlib, a framework for generating spatial hardware accelerators for tensor algebra applications. Tensorlib is motivated by the observation that, tensor dataflows can be expressed with linear transformations, and they share common hardware modules which can be reused across different designs. Tensorlib first uses Space-Time Transformation to explore different dataflows, which can compactly represent the hardware dataflow using a transformation matrix. Next, we identify the common structures of different dataflows and build parameterized hardware module templates. Our generation framework can select the needed hardware modules for each dataflow, connect the modules using a specified interconnection pattern, and automatically generate the complete hardware accelerator design. Tensorlib remarkably improves the productivity for the development and optimization of spatial hardware architecture, providing a rich design space with trade-offs in performance, area, and power. Experiments show that Tensorlib can automatically generate hardware designs with different dataflows for a variety of tensor algebra programs. Tensorlib can achieve 318 MHz frequency and 786 GFLOP/s throughput for matrix multiplication kernel on Xilinx VU9P FPGA, which outperforms the state-of-the-art generators.
- 27 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (6) , 1898-1911
Open AccessAutomatic Generation of Spatial Accelerator for Tensor Algebra
Tensor algebra finds applications in various domains including machine learning applications, data analytics and others. Spatial hardware accelerators are widely used to boost the performance of tensor algebra applications. It has a complex hardware architecture and rich design space. Prior approaches based on manual implementation lead to low programming productivity, making it hard to explore the large design space. In this paper, we propose Tensorlib, a framework for generating spatial hardware accelerators for tensor algebra applications. Tensorlib is motivated by the observation that, tensor dataflows can be expressed with linear transformations, and they share common hardware modules which can be reused across different designs. Tensorlib first uses Space-Time Transformation to explore different dataflows, which can compactly represent the hardware dataflow using a transformation matrix. Next, we identify the common structures of different dataflows and build parameterized hardware module templates. Our generation framework can select the needed hardware modules for each dataflow, connect the modules using a specified interconnection pattern, and automatically generate the complete hardware accelerator design. Tensorlib remarkably improves the productivity for the development and optimization of spatial hardware architecture, providing a rich design space with trade-offs in performance, area, and power. Experiments show that Tensorlib can automatically generate hardware designs with different dataflows for a variety of tensor algebra programs. Tensorlib can achieve 318 MHz frequency and 786 GFLOP/s throughput for matrix multiplication kernel on Xilinx VU9P FPGA, which outperforms the state-of-the-art generators.
- Open Access
- 26 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 40 (4) , 85-92
Generating Systolic Array Accelerators With Reusable Blocks
Systolic array architecture is widely used in spatial hardware and well-suited for many tensor applications. Many systolic array architectures are implemented with high level synthesis (HLS) design flow. However, existing HLS tools do not favor of modular and reusable design, which brings inefficiency for hardware design. In this paper, we analyze the systolic array design space, and identify the common structures of different systolic dataflows. We build hardware module templates using Chisel infrastructure, which can be reused for different dataflows. This can remarkably improve the productivity of systolic array based architecture design. We further build a systolic array generator that transforms the tensor algorithm definition to a complete systolic hardware architecture. Experiments show that we can implement systolic array designs for different applications and dataflows with little engineering effort, and the performance throughput outperforms HLS designs.
- 26 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 40 (4) , 85-92
Open AccessGenerating Systolic Array Accelerators With Reusable Blocks
Systolic array architecture is widely used in spatial hardware and well-suited for many tensor applications. Many systolic array architectures are implemented with high level synthesis (HLS) design flow. However, existing HLS tools do not favor of modular and reusable design, which brings inefficiency for hardware design. In this paper, we analyze the systolic array design space, and identify the common structures of different systolic dataflows. We build hardware module templates using Chisel infrastructure, which can be reused for different dataflows. This can remarkably improve the productivity of systolic array based architecture design. We further build a systolic array generator that transforms the tensor algorithm definition to a complete systolic hardware architecture. Experiments show that we can implement systolic array designs for different applications and dataflows with little engineering effort, and the performance throughput outperforms HLS designs.
- Open Access
- 21 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. PP (00189340) , 1
Enabling Efficient Fast Convolution Algorithms on GPUs via MegaKernels
Modern CNNs require huge amount of convolution operations. To address the overwhelming computation problem, Winograd and FFT fast algorithms have been used as effective approaches to reduce the number of multiplications. Inputs and filters are transformed to special domains then perform element-wise multiplication, which can be transformed into batched GEMM operation. Different stages of computation contain multiple tasks with different computation and memory behaviors, and they share intermediate data. This provides the opportunity to fuse these tasks into a monolithic kernel. In this paper, we propose a kernel fusion technique for fast convolution algorithms based on MegaKernels. Each GPU thread block is assigned with one computation stage and we design a parameterized task mapping algorithm to assign stages to thread blocks. We build a task scheduler which fetches and executes the tasks following the task dependency. Evaluation on modern CNNs shows that our techniques achieve an average of 1.25X and 1.7X speedup compared to cuDNN's two implementations on Winograd convolution algorithm.
- 21 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. PP (00189340) , 1
Open AccessEnabling Efficient Fast Convolution Algorithms on GPUs via MegaKernels
Modern CNNs require huge amount of convolution operations. To address the overwhelming computation problem, Winograd and FFT fast algorithms have been used as effective approaches to reduce the number of multiplications. Inputs and filters are transformed to special domains then perform element-wise multiplication, which can be transformed into batched GEMM operation. Different stages of computation contain multiple tasks with different computation and memory behaviors, and they share intermediate data. This provides the opportunity to fuse these tasks into a monolithic kernel. In this paper, we propose a kernel fusion technique for fast convolution algorithms based on MegaKernels. Each GPU thread block is assigned with one computation stage and we design a parameterized task mapping algorithm to assign stages to thread blocks. We build a task scheduler which fetches and executes the tasks following the task dependency. Evaluation on modern CNNs shows that our techniques achieve an average of 1.25X and 1.7X speedup compared to cuDNN's two implementations on Winograd convolution algorithm.