Chen Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Chen Chen has a total of 100 co-authors. Here are the top co-authors by number of shared publications:
- Minyi Guo
Shanghai Jiao Tong University, Shanghai, China
17 shared publications - Quan Chen
Shanghai Jiao Tong University
12 shared publications - Bo Li
The Hong Kong University of Science and Technology
5 shared publications - Jiayi Zhang
Shanghai Jiao Tong University, Shanghai, China
5 shared publications - Jieru Zhao
Shanghai Jiao Tong University, Shanghai, China
4 shared publications - Jingwen Leng
Shanghai Jiao Tong University, Shanghai, China
4 shared publications
Latest Publications
- Open Access
- 10 April 2027
- Published by Association for Computing Machinery (ACM)
- p. 554-572
Impeller: Scalable and Efficient Zero-Knowledge Proof Computation Framework via Graph Compilation
- 10 April 2027
- Published by Association for Computing Machinery (ACM)
- p. 554-572
Open AccessImpeller: Scalable and Efficient Zero-Knowledge Proof Computation Framework via Graph Compilation
- Open Access
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 45-55
Aethon: Performance-aware Memory Offloading for Co-running Applications in Public Clouds
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 45-55
Open AccessAethon: Performance-aware Memory Offloading for Co-running Applications in Public Clouds
- Open Access
- 5 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 260-266
Towards Efficient Serving of Network-intensive LLM Inferences
- 5 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 260-266
Open AccessTowards Efficient Serving of Network-intensive LLM Inferences
- Open Access
- 13 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 564-566
FluxZK: Scalable and Efficient Zero-Knowledge Proof Computation via GPU Acceleration
Zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs) are a key technology to privacy-preserving applications today. The complexity of proof generation, however, heavily constrains throughput in latency-sensitive environments. The computational burden primarily stems from two fundamental algorithms: Multi-Scalar Multiplication (MSM) and the Number Theoretic Transform (NTT). We propose a series of optimizations for these two kernels, including computation-transfer pipelining, load balancing, and memory access fusion, achieving 1.97 × to 2.16 × proof generation speedup over a state-of-the-art open source GPU acceleration library. Our design also supports out-of-core computation, enabling the generation of large-scale ZKP proofs.
- 13 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 564-566
Open AccessFluxZK: Scalable and Efficient Zero-Knowledge Proof Computation via GPU Acceleration
Zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs) are a key technology to privacy-preserving applications today. The complexity of proof generation, however, heavily constrains throughput in latency-sensitive environments. The computational burden primarily stems from two fundamental algorithms: Multi-Scalar Multiplication (MSM) and the Number Theoretic Transform (NTT). We propose a series of optimizations for these two kernels, including computation-transfer pipelining, load balancing, and memory access fusion, achieving 1.97 × to 2.16 × proof generation speedup over a state-of-the-art open source GPU acceleration library. Our design also supports out-of-core computation, enabling the generation of large-scale ZKP proofs.
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 37  (7) , 1589-1604
Enabling Client-Autonomous Training Optimizations for Efficient Federated Learning
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 37  (7) , 1589-1604
Enabling Client-Autonomous Training Optimizations for Efficient Federated Learning
- Open Access
- 27 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23  (2) , 1-25
Hermes: Efficient Serving of LLM Applications with Probabilistic Demand Modeling
Applications based on Large Language Models (LLMs) contain a series of tasks to address real-world problems with boosted capability, which have dynamic demand volumes on diverse backends. Existing serving systems treat the resource demands of LLM applications as a blackbox, compromising end-to-end efficiency due to improper queuing order and backend warm up latency. We find that the resource demands of LLM applications can be modeled in a general and accurate manner with Probabilistic Demand Graph (PDGraph). We then propose Hermes, which leverages PDGraph for efficient serving of LLM applications. Confronting probabilistic demand description, Hermes applies the Gittins policy to determine the scheduling order that can minimize the average application completion time. It also uses the PDGraph model to help prewarm cold backends at proper moments. Experiments with diverse LLM applications confirm that Hermes can effectively improve the application serving efficiency, reducing the average completion time by over 70% and the P95 completion time by over 80%.
- 27 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23  (2) , 1-25
Open AccessHermes: Efficient Serving of LLM Applications with Probabilistic Demand Modeling
Applications based on Large Language Models (LLMs) contain a series of tasks to address real-world problems with boosted capability, which have dynamic demand volumes on diverse backends. Existing serving systems treat the resource demands of LLM applications as a blackbox, compromising end-to-end efficiency due to improper queuing order and backend warm up latency. We find that the resource demands of LLM applications can be modeled in a general and accurate manner with Probabilistic Demand Graph (PDGraph). We then propose Hermes, which leverages PDGraph for efficient serving of LLM applications. Confronting probabilistic demand description, Hermes applies the Gittins policy to determine the scheduling order that can minimize the average application completion time. It also uses the PDGraph model to help prewarm cold backends at proper moments. Experiments with diverse LLM applications confirm that Hermes can effectively improve the application serving efficiency, reducing the average completion time by over 70% and the P95 completion time by over 80%.
- Open Access
- 26 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23  (2) , 1-26
NPUMeter: Automatic Operator Optimization for Ascend NPU with Accurate Analytical Performance Models
With the rapid development of AI and deep learning, computational demands are increasing significantly. While GPUs excel in parallel computing, they fall short in terms of energy efficiency, specialization, and processing latency. In contrast, Neural Processing Units (NPUs), such as the Ascend NPUs, designed specifically for deep learning tasks, demonstrate superior performance. However, the architecture specialization makes operator development more challenging, leading to a reliance on manual tuning and optimization, which incurs significant time cost and developing effort. To address this issue, we propose NPUMeter, an automatic operator optimization framework for Ascend NPUs built upon accurate and comprehensive analytical performance models. NPUMeter comprises two components: (1) an analytical performance model that accurately estimates operator latency on NPU given different configurations of optimization parameters; (2) an efficient design space exploration (DSE) algorithm that automatically searches for the optimal parameter configuration in a large design space within minutes. Experimental results demonstrate that NPUMeter achieves high estimation accuracy, with an average error below 5%. It effectively generates near-optimal configurations for various operators, achieving up to a 1.46× performance speedup compared to the configuration generated by the Ascend C compiler while reducing the DSE time from hours to minutes.
- 26 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23  (2) , 1-26
Open AccessNPUMeter: Automatic Operator Optimization for Ascend NPU with Accurate Analytical Performance Models
With the rapid development of AI and deep learning, computational demands are increasing significantly. While GPUs excel in parallel computing, they fall short in terms of energy efficiency, specialization, and processing latency. In contrast, Neural Processing Units (NPUs), such as the Ascend NPUs, designed specifically for deep learning tasks, demonstrate superior performance. However, the architecture specialization makes operator development more challenging, leading to a reliance on manual tuning and optimization, which incurs significant time cost and developing effort. To address this issue, we propose NPUMeter, an automatic operator optimization framework for Ascend NPUs built upon accurate and comprehensive analytical performance models. NPUMeter comprises two components: (1) an analytical performance model that accurately estimates operator latency on NPU given different configurations of optimization parameters; (2) an efficient design space exploration (DSE) algorithm that automatically searches for the optimal parameter configuration in a large design space within minutes. Experimental results demonstrate that NPUMeter achieves high estimation accuracy, with an average error below 5%. It effectively generates near-optimal configurations for various operators, achieving up to a 1.46× performance speedup compared to the configuration generated by the Ascend C compiler while reducing the DSE time from hours to minutes.
- Open Access
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2002-2021
Suika: Efficient and High-quality Re-scheduling of 3D-parallelized LLM Training Jobs in Shared Clusters
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2002-2021
Open AccessSuika: Efficient and High-quality Re-scheduling of 3D-parallelized LLM Training Jobs in Shared Clusters
- Open Access
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 89-107
Arena: Efficiently Training Large Models via Dynamic Scheduling and Adaptive Parallelism Co-Design
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 89-107
Open AccessArena: Efficiently Training Large Models via Dynamic Scheduling and Adaptive Parallelism Co-Design
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP  (15361233) , 1-16
Boosting Gradient-based Training Diagnosis for Efficient and Accurate Federated Learning
Federated Learning (FL) allows edge clients to collaborate in model training with data privacy preserved, yet it is known to suffer low training efficiency and model accuracy. Given that efficiency and accuracy are usually conflicting objectives, existing practices increasingly employ an adaptive scheme that changes the FL configurations (e.g., quantization or sparsification level) based on runtime training status, for which accurate training diagnosis—used for guiding the optimization actions—is crucial. However, while training diagnosis is a common task shared by different optimization schemes, existing works propose their diagnosis methods in an ad-hoc manner, which yield multiple limitations. First, the diagnosis metric in an optimization scheme may sometimes be less accurate than others; second, existing schemes fail to fully exploit the diagnosis result by applying it for only one optimization action; third, existing methods usually do not perceive cross-client data heterogeneity, failing to simultaneously enhance FL accuracy. To tackle those limitations, we make a systematical study on the training diagnosis methods of multiple optimization schemes, and propose metric grafting—replacing a scheme's diagnosis metric with a better one to improve the training performance. Moreover, to fully exploit the potential of training diagnosis, we build a system platform that supports flexible combinations of training diagnosis and optimization actions (i.e., single-diagnosis-multiple actions and multiple-diagnosis-multiple-actions). Evaluation on testbeds show that, with metric grafting and advanced diagnosis action combinations, we can substantially improve the efficiency and accuracy performance of FL.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP  (15361233) , 1-16
Boosting Gradient-based Training Diagnosis for Efficient and Accurate Federated Learning
Federated Learning (FL) allows edge clients to collaborate in model training with data privacy preserved, yet it is known to suffer low training efficiency and model accuracy. Given that efficiency and accuracy are usually conflicting objectives, existing practices increasingly employ an adaptive scheme that changes the FL configurations (e.g., quantization or sparsification level) based on runtime training status, for which accurate training diagnosis—used for guiding the optimization actions—is crucial. However, while training diagnosis is a common task shared by different optimization schemes, existing works propose their diagnosis methods in an ad-hoc manner, which yield multiple limitations. First, the diagnosis metric in an optimization scheme may sometimes be less accurate than others; second, existing schemes fail to fully exploit the diagnosis result by applying it for only one optimization action; third, existing methods usually do not perceive cross-client data heterogeneity, failing to simultaneously enhance FL accuracy. To tackle those limitations, we make a systematical study on the training diagnosis methods of multiple optimization schemes, and propose metric grafting—replacing a scheme's diagnosis metric with a better one to improve the training performance. Moreover, to fully exploit the potential of training diagnosis, we build a system platform that supports flexible combinations of training diagnosis and optimization actions (i.e., single-diagnosis-multiple actions and multiple-diagnosis-multiple-actions). Evaluation on testbeds show that, with metric grafting and advanced diagnosis action combinations, we can substantially improve the efficiency and accuracy performance of FL.