Chunshu Wu

Chunshu Wu

Auto-generated profile, based on publicly available information

Chunshu Wu

Pacific Northwest National Laboratory

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
21 June 2023
Verified
Conference Paper
Open access
AGAnqi Guo
Anqi Guo
YHYuchen Hao
Yuchen Hao
CWChunshu Wu
Chunshu Wu
PHPouya Haghi
Pouya Haghi
ZPZhenyu Pan
Zhenyu Pan
TGTong Geng
Tong Geng
Deep Learning Recommendation Models (DLRMs) are important applications in various domains and have evolved into one of the largest and most important machine learning applications. With their trillions of parameters necessarily exceeding the high bandwidth memory (HBM) capacity of GPUs, ever more massive DLRMs require large-scale multi-node systems for distributed training and inference. However, these all suffer from the all-to-all communication bottleneck, which limits scalability. SmartNICs couple computation and communication capabilities to provide powerful network-facing heterogeneous devices that reduce communication overhead. There has not, however, been a distributed system design that fully leverages SmartNIC resources to address scalability of DLRMs. We propose a software-hardware co-design of a heterogeneous SmartNIC system that overcomes the communication bottleneck of distributed DLRMs, mitigates the pressure on memory bandwidth, and improves computation efficiency. We provide a set of SmartNIC designs of cache systems (including local cache and remote cache) and SmartNIC computation kernels that reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches and optimizes the overall system performance with higher data reuse. Our evaluation shows that the system achieves 2.1× latency speedup for inference and 1.6× throughput speedup for training.

11 November 2023
Verified
Conference Paper
Open access
CWChunshu Wu
Chunshu Wu
TGTong Geng
Tong Geng
AGAnqi Guo
Anqi Guo
SBSahan Bandara
Sahan Bandara
PHPouya Haghi
Pouya Haghi
MHMartin Herbordt
Martin Herbordt
Conducting long-timescale simulations of small molecules using Molecular Dynamics (MD) is crucial in drug design. However, traditional methods to accelerate the process, including ASICs or GPUs, have limitations. ASIC solutions are not always generally available, while GPU solutions may not scale when processing small molecules. FPGAs are both communication processors and accelerators, with tight coupling between these capabilities, and so could be used to address strong scaling in this domain. We present FASDA, the first FPGA-based MD accelerator available for community development. FASDA enables the use of FPGA enhanced clusters and clouds to execute range-limited MD, which is the most resource-intensive and computation-demanding component in MD. FASDA is built with a series of plugable components that are adjustable based on user requirements and demonstrates nearly linear scaling on an eight FPGA cluster. It outperforms the state-of-the-art GPU solution by 4.67x, with the resulting prospect of significantly reducing lead evaluation time.

21 June 2023
Verified
Conference Paper
Open access
PHPouya Haghi
Pouya Haghi
WKWilliam Krska
William Krska
CTCheng Tan
Cheng Tan
TGTong Geng
Tong Geng
PCPo Hao Chen
Po Hao Chen
MHMartin Herbordt
Martin Herbordt
Some communication switches, e.g., the Mellanox SHArP and those in the IBM BlueGene clusters, are augmented to process packets at the application level with fixed-function collectives. This approach, however, lacks flexibility, which limits their applicability in diverse and dynamic workloads. Recently, a new type of programmable packet processor, which uses high-level languages, e.g., P4, has emerged as a possible candidate. P4-based switches, however, fall short in certain applications, including machine learning, where capabilities not currently supported by P4 are needed. These include more complex calculation, such as sparse computation and fused multiply-accumulate, data-intensive floating point operations, data reuse, and significant memory. The problem addressed here is that such a switch augmentation needs to support: a large amount of state, significant flexible compute capability, and ease of programming, all while maintaining full functionality, including ensuring high throughput, and demonstrating utility. In this work, we propose a programmable look-aside-type accelerator that can be embedded into, or attached to, existing communication switch pipelines and that is capable of processing packets at line rate. The proposed in-switch accelerator is based on mixing an ISA (subset of RISC-V instructions) with dataflow graphs (found in CGRAs). To augment performance, vector instructions are also supported. To facilitate usability, we have developed a complete toolchain to compile user-provided C/C++ codes to appropriate back-end instructions for configuring the accelerator. While this approach is flexible enough to support various workloads, in this paper, we consider Graph Convolutional Networks (GCNs) as a case study. Experimental results show that this approach considerably improves the performance of distributed GCN applications.

30 May 2024
Verified
Conference Paper
Open access
PHPouya Haghi
Pouya Haghi
CTCheng Tan
Cheng Tan
AGAnqi Guo
Anqi Guo
CWChunshu Wu
Chunshu Wu
Dongfang LiuDongfang Liu
Dongfang Liu
MHMartin Herbordt
Martin Herbordt
Communication switches have sometimes been augmented to process collectives, e.g., in the IBM BlueGene and Mellanox SHArP switches. In this work, we find that there is a great acceleration opportunity through the further augmentation of switches to accelerate more complex functions that combine communication with computation. We consider three types of such functions. The first is fully-fused collectives built by fusing multiple existing collectives like Allreduce with Alltoall. The second is semi-fused collectives built by combining a collective with another computation. The third are higher-order collectives built by combining multiple computations and communications, such as to perform matrix-matrix multiply (PGEMM). In this work, we propose a framework called SmartFuse to accelerate fused collective functions. The core of SmartFuse is a reconfigurable smart switch to support these operations. The semi/fully fused collectives are implemented with a CGRA-like architecture, while higher-order collectives are implemented with a more specialized computational unit that can also schedule communication. Supporting our framework is software to evaluate and translate relevant parts of the input program, compile them into a control data flow graph, and then map this graph to the switch hardware. The proposed framework, once deployed, has the strong potential to accelerate existing HPC applications transparently by encapsulation within an MPI implementation. Experimental results show that this approach improves the performance of the PGEMM kernel, miniFE, and AMG by, on average, 94%, 15%, and 13%, respectively.

14 March 2024 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
CWChunshu Wu
Chunshu Wu
CYChen Yang
Chen Yang
SBSahan Bandara
Sahan Bandara
TGTong Geng
Tong Geng
AGAnqi Guo
Anqi Guo
MHMartin Herbordt
Martin Herbordt
Long timescale Molecular Dynamics (MD) simulation of small molecules is crucial in drug design and basic science. To accelerate a small data set that is executed for a large number of iterations, high-efficiency is required. Recent work in this domain has demonstrated that among COTS devices only FPGA-centric clusters can scale beyond a few processors. The problem addressed here is that, as the number of on-chip processors has increased from fewer than 10 into the hundreds, previous intra-chip routing solutions are no longer viable. We find, however, that through various design innovations, high efficiency can be maintained. These include replacing the previous broadcast networks with ring-routing and then augmenting the rings with out-of-order and caching mechanisms. Others are adding a level of hierarchical filtering and memory recycling. Two novel optimized architectures emerge, together with a number of variations. These are validated, analyzed, and evaluated. We find that in the domain of interest speed-ups over GPUs are achieved. The potential impact is that this system promises to be the basis for scalable long timescale MD with commodity clusters.

17 October 2025
Verified
Conference Paper
Open access
CLChuan Liu
Chuan Liu
CWChunshu Wu
Chunshu Wu
RSRuibing Song
Ruibing Song
GSGuangyan Sun
Guangyan Sun
YWYingnian Wu
Yingnian Wu
TGTong Geng
Tong Geng