ZWZiheng Wang

Ziheng Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ziheng Wang has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    TSE-IO: Two-Stage Extrapolative I/O Modeling and Tuning via Pairwise Learning

    by Heng Liu, Xiaoshe Dong, Qiang Wang, Yifan Li, XinDong Liu, BingZhi Long, Ziheng Wang

    Collecting I/O data for parallel programs at large node scales is prohibitively expensive and has long been a major bottleneck for I/O performance modeling. Existing approaches typically depend on extensive training samples collected on large-scale systems, which incur high cost and generalize poorly across applications and environments. To address these challenges, we propose TSE-IO, a T wo- S tage framework for E xtrapolative I/O performance modeling and tuning. TSE-IO decomposes cross-scale performance prediction into two linked tasks: first, extrapolating application-level I/O behavior from small-scale measurements to the target scale. And second, predicting target-scale performance by jointly modeling the extrapolated behavior and tunable I/O-stack configurations. For the second stage, TSE-IO introduces a pairwise formulation that combines delta-feature matching with anchor-based inference to transfer configuration-induced performance patterns learned at small scales to unseen target scales. Experimental results show that, when trained solely on measurements collected from 1-16 nodes, TSE-IO extrapolates the I/O time of IOR, S3D-IO, Flash-IO, FLASH, and LAMMPS to 128 nodes (2,048 processes), achieving a MAPE below 20% for the majority of evaluated workloads. After tuning, five representative applications exhibit significant I/O performance improvements, achieving speedups ranging from 16.38 × to 65.38 ×, and from 1.00 × to 2.79 × compared with expert user settings across different workloads, thereby demonstrating the effectiveness and generality of TSE-IO.

  • Open Access

    BLG-Tuning: Benchmark-Based Low-Cost General-Purpose I/O Modeling and Tuning

    by Ziheng Wang, Chaoqun Zhang, Yifan Li, Yuchao Wu, Yuping Wang, Xiaoshe Dong

    I/O performance has become a major bottleneck for many data-intensive applications. Each layer of the parallel I/O stack provides parameters that can optimize I/O performance, but determining the optimal performance parameters based on the operating configuration is a challenge. Previous work has required separate performance models for different programs for tuning, which is very costly in term of measurement data. We propose BLG-Tuning: a B enchmark-based L ow-cost G eneral-purpose I/O Modeling and Tuning. BLG-Tuning maps application I/O loads to benchmark parameters and uses the benchmark-trained performance model to achieve I/O performance prediction and thus avoid the additional computing and communication overhead for measurement. For applications, BLG-Tuning collects the application characteristics to calibrate the performance model and improve prediction accuracy. Experience shows that BLG-Tuning predicts the I/O time of MADbench2, Flash-IO, S3D-IO, BT-IO, and LAMMPS with MAPE of 22.2%, 18.2%, 29.3%, 21.5%, and 34.8%, respectively. After tuning, the five applications obtain I/O speedup from 5.6× to 27.3×.

  • Mapo: Performance model driven GPU memory access code optimization

    by Shuai Liu, Xiaoshe Dong, Junkai Cao, Ruifan Chu, Ziheng Wang, Qiang Wang, Xiuxiu Bai

  • Open Access

    A Portable GPU Kernel Performance Modeling Method Based on LLVM IR Dynamic Feature Prediction

    by Qiang Wang, Hao Zheng, Yiru Liu, Fan Yang, Jiachen Liu, Chaojun Deng, Ziheng Wang, Xiaoshe Dong

  • Open Access

    An Example of Parallel Merkle Tree Traversal: Post-Quantum Leighton-Micali Signature on the GPU

    by Ziheng Wang, Xiaoshe Dong, Yan Kang, Heng Chen, Qiang Wang

    The hash-based signature (HBS) is the most conservative and time-consuming among many post-quantum cryptography (PQC) algorithms. Two HBSs, LMS and XMSS, are the only PQC algorithms standardised by the National Institute of Standards and Technology (NIST) now. Existing HBSs are designed based on serial Merkle tree traversal, which is not conducive to taking full advantage of the computing power of parallel architectures such as CPUs and GPUs. We propose a parallel Merkle tree traversal (PMTT), which is tested by implementing LMS on the GPU. This is the first work accelerating LMS on the GPU, which performs well even with over 10,000 cores. Considering different scenarios of algorithmic parallelism and data parallelism, we implement corresponding variants for PMTT. The design of PMTT for algorithmic parallelism mainly considers the execution efficiency of a single task, while that for data parallelism starts with the full utilisation of GPU performance. In addition, we are the first to design a CPU-GPU collaborative processing solution for traversal algorithms to reduce the communication overhead between CPU and GPU. For algorithmic parallelism, our implementation is still 4.48× faster than the ideal time of the state-of-the-art traversal algorithm. For data parallelism, when the number of cores increases from 1 to 8,192, the parallel efficiency is 78.39%. In comparison, our LMS implementation outperforms most existing LMS and XMSS implementations.


  • CUSPX: Efficient GPU Implementations of Post-Quantum Signature SPHINCS +

    by Ziheng Wang, Xiaoshe Dong, Heng Chen, Yan Kang, Qiang Wang

    Quantum computers pose a serious threat to existing cryptographic systems. While Post-Quantum Cryptography (PQC) offers resilience against quantum attacks, its performance limitations often hinder widespread adoption. Among the three National Institute of Standards and Technology (NIST)-selected general-purpose PQC schemes, SPHINCS ${}^{+}$ is particularly susceptible to these limitations. We introduce CUSPX ( CU DA SP HIN CS ${}^{+}$ ), the first large-scale parallel implementation of SPHINCS ${}^{+}$ capable of running across 10,000 cores. CUSPX leverages a novel three-level parallelism framework, applying it to algorithmic parallelism , data parallelism , and hybrid parallelism . Notably, CUSPX introduces parallel Merkle tree construction algorithms for arbitrary parallel scales and several load-balancing solutions, further enhancing performance. By treating tasks parallelism as the top level of parallelism, CUSPX provides a four-level parallel scheme that can run with any number of tasks. Evaluated on a single GeForce RTX 3090 using the SPHINCS ${}^{+}$ -SHA-256-128s-simple parameter set, CUSPX achieves a single task's signature generation latency of 0.67 ms, demonstrating a 5,105 $\times$ speedup over a single-thread version and an 18.50 $\times$ speedup over the previous fastest implementation.


  • Parallel implementations of post-quantum leighton-Micali signature on multiple nodes

    by Yan Kang, Xiaoshe Dong, Ziheng Wang, Heng Chen, Qiang Wang


  • Open Access

    Parallel Implementations of Post-Quantum Leighton-Micali Signature on Multiple Nodes

    by Yan Kang, Xiaoshe Dong, Ziheng Wang, Heng Chen, Qiang Wang

    To defend against quantum computer attacks, the National Institute of Standards and Technology (NIST) has been exploring post-quantum cryptography (PQC). Now, NIST has standardised only two PQC algorithms, one of which is the Leighton-Micali signature (LMS). However, the performance of LMS limits its practical application. In this paper, we propose a parallel LMS implementation on multiple nodes. Considering different application scenarios, we provide two parallel schemes: algorithmic parallelism and data parallelism. The main part of our work is the two-tier parallel structure for the LMS tree. Targeting the x86/64 multiple nodes, our work introduces vectorization to present the three-tier parallel structure. We also design communication optimization, including the selection of communication primitives and the creation of communicators for multi-node running. Experimental evidence shows that our code effectively reduces the latency, and is 19.04× faster than the fastest implementation on the same platform when running key pair generation for LMS SHA256 M32 H20(20).


  • Efficient GPU Implementations of Post-Quantum Signature XMSS

    by Ziheng Wang, Xiaoshe Dong, Heng Chen, Yan Kang

    The National Institute of Standards and Technology (NIST) approved XMSS as part of the post-quantum cryptography (PQC) development effort in 2018. XMSS is currently one of only two standardized PQC algorithms, but its performance limits its use. For example, the fastest record for some standardized parameters still takes more than a minute to generate a keypair. In this article, we present the first GPU implementation for XMSS and its variant XMSS $^{\mathsf {MT}}$ . The high parallelism of GPUs is especially effective for reducing latency in key generation and improving throughput for signing and verifying. In order to meet various application scenarios, we provide three parallel XMSS schemes: algorithmic parallelism , multi-keypair data parallelism , and single-keypair data parallelism . For these schemes, we design custom parallel strategies that use more than 10,000 cores for all parameters provided by NIST. In addition, we analyze the availability of most previous serial optimizations and explore numerous techniques to fully exploit GPU performance. Our evaluations are made with the XMSSMT-SHA2_20/2_256 parameter set on a GeForce RTX 3090. The result shows the key generation latency is 3.20 ms, a speedup of 21,899× compared to the GPU ported version, which is also 54× speedup faster than the fastest work (174 ms). When 16384 tasks are executed, the throughput (task/s) for signing/verifying in the single-key and multi-key cases is 311,424/415,100 and 145,100/419,887, respectively. Compared to the throughput for signing/verifying (1695/4000) of the fastest work, we obtain a speedup of 184×/104× and 86×/105× in single-key and multi-key cases, respectively.