Mingzhe Zhang

Mingzhe Zhang

Auto-generated profile, based on publicly available information

Mingzhe Zhang

Tsinghua University, Beijing, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 April 2024
Verified
Conference Paper
Open access
XHXiaochen Hao
Xiaochen Hao
HRHongbo Rong
Hongbo Rong
MZMingzhe Zhang
Mingzhe Zhang
CSCe Sun
Ce Sun
HJHong Jiang
Hong Jiang
YLYun Liang
Yun Liang
This paper aims at high and portable performance for tensor computations across spatial (e.g., FPGAs) and vector architectures (e.g., GPUs). The state-of-the-art usually address performance portability across vector architectures (CPUs and GPUs). However, they either miss FPGAs or do not achieve high performance. Without a common architectural abstraction, they program and optimize spatial and vector devices separately, causing low portability. We propose a unified programming framework, POPA, which achieves portability via architectural abstraction and performance via specialization. A parallel dataflow machine is proposed as a unified, abstract hardware target that hides differences of concrete architectures. The machine consists of software-defined systolic arrays and a tensor-specific cache hierarchy, which captures pipeline parallelism and customizable memories on FPGAs, as well as multithreading parallelism on GPUs. The machine is specified in a unified programming model as two dataflow graphs for scheduling compute and data movement, respectively. A compiler then specializes the abstract machine to exploit the properties of FPGAs and GPUs, bridging the gap between the abstract machine and a concrete architecture. We evaluate POPA on several Intel FPGAs and GPUs with high-profile tensor kernels, and this is the first system that achieves >=80% performance of expert-written code or machine peak across architectures, to the best of our knowledge.

31 May 2025 in ACM Transactions on Reconfigurable Technology and Systems
Verified
Journal Article
Open access
XHXiaochen Hao
Xiaochen Hao
MZMingzhe Zhang
Mingzhe Zhang
CSCe Sun
Ce Sun
ZTZhuofu Tao
Zhuofu Tao
HRHongbo Rong
Hongbo Rong
YLYun Liang
Yun Liang
Linear algebra computations can be greatly accelerated using spatial accelerators on FPGAs. As a standard building block of linear algebra applications, BLAS covers a wide range of compute patterns that vary vastly in data reuse, bottleneck resources, matrix storage layouts, and data types. However, existing implementations of BLAS routines on FPGAs are stuck in the dilemma of productivity and performance. They either require extensive human effort or fail to leverage the properties of routines for acceleration. We introduce Lasa, a framework composed of a programming model and a compiler, designed to address the dilemma by abstracting (for productivity) and specializing (for performance) the architecture of a spatial accelerator. The programming model realizes systolic arrays using uniform recurrence equations and space-time transforms. Streaming tensors, an intuitive dataflow-style abstraction, is proposed to uniformly describe the movement, storage, and transpose of input and output data across the spatial components. According to streaming tensors, a customized memory hierarchy is automatically built on an FPGA by our compiler. The compiler further specializes the architecture with transparent optimizations on FPGAs. Using this framework, we develop a complete BLAS library, demonstrating performance in parity with expert-written HLS code for BLAS level 3 routines, 76%–94% machine peak for level 1 and 2 routines, and 1.6X–13X speedup by leveraging the matrix properties such as symmetry, triangularity, and bandness.

27 October 2024
Verified
Conference Paper
Open access
MZMingzhe Zhang
Mingzhe Zhang
XHXiaochen Hao
Xiaochen Hao
HRHongbo Rong
Hongbo Rong
WCWenguang Chen
Wenguang Chen