Page 224 - 《软件学报》2026年第6期

周文浩 等: SWTuner: 基于机器学习方法的分布式编译调优框架 2543

未来的研究将深入探讨 SWTuner 在更大规模并行计算环境中的应用, 特别是在异构计算平台上的性能表现.
同时, 本研究将致力于提高模型的预测精度, 特别是在处理新型应用或架构时, 增强模型的泛化能力. 此外, 还将探
索更高效的搜索策略, 以更好地平衡探索与利用之间的关系, 避免陷入局部最优解. 最后, 计划将 SWTuner 扩展至
支持更多的编译器和编程模型, 使其成为一个更为广泛适用的自动编译调优工具.

References
[1] Bacon DF, Graham SL, Sharp OJ. Compiler transformations for high-performance computing. ACM Computing Surveys (CSUR), 1994,
26(4): 345–420. [doi: 10.1145/197405.197406]
[2] Sarkar V. Optimized unrolling of nested loops. In: Proc. of the 14th Int’l Conf. on Supercomputing. New Mexico: ACM, 2000. 153–166.
[doi: 10.1145/335231.335246]
[3] Wegman MN, Zadeck FK. Constant propagation with conditional branches. ACM Trans. on Programming Languages and Systems
(TOPLAS), 1991, 13(2): 181–210. [doi: 10.1145/103135.103136]
[4] Golovanevsky O, Zaks A. Struct-reorg: Current status and future perspectives. In: Proc. of the 2007 GCC Developers’ Summit. 2007.
47–56.
[5] Ayers A, Schooler R, Gottlieb R. Aggressive inlining. In: Proc. of the 1997 ACM SIGPLAN Conf. on Programming Language Design
and Implementation. Las Vegas: ACM, 1997. 134–145. [doi: 10.1145/258915.258928]
[6] Nuzman D, Rosen I, Zaks A. Auto-vectorization of interleaved data for SIMD. ACM SIGPLAN Notices, 2006, 41(6): 132–143. [doi: 10.
1145/1133255.1133997]
[7] Dagum L, Menon R. OpenMP: An industry standard API for shared-memory programming. IEEE Computational Science and
Engineering, 1998, 5(1): 46–55. [doi: 10.1109/99.660313]
[8] Ryoo S, Rodrigues CI, Baghsorkhi SS, Stone SS, Kirk DB, Hwu WMW. Optimization principles and application performance evaluation
of a multithreaded GPU using CUDA. In: Proc. of the 13th ACM SIGPLAN Symp. on Principles and Practice of Parallel Programming.
Salt Lake City: ACM, 2008. 73–82. [doi: 10.1145/1345206.1345220]
[9] Chung IH, Hollingsworth JK. A case study using automatic performance tuning for large-scale scientific programs. In: Proc. of the 15th
IEEE Int’l Conf. on High Performance Distributed Computing. Paris: IEEE, 2006. 45–56. [doi: 10.1109/HPDC.2006.1652135]
[10] Reinders J, Ashbaugh B, Brodman J, Kinsner M, Pennycook J, Tian XM. Data Parallel C++: Mastering DPC++ for Programming of
Heterogeneous Systems Using C++ and SYCL. New York: Apress Berkeley, 2021. [doi: 10.1007/978-1-4842-5574-2]
[11] Evans AN, Campbell B, Soffa ML. Is Rust used safely by software developers? In: Proc. of the 42nd ACM/IEEE Int’l Conf. on Software
Engineering. Seoul: ACM, 2020. 246–257. [doi: 10.1145/3377811.3380413]
[12] Bezanson J, Edelman A, Karpinski S, Shah VB. Julia: A fresh approach to numerical computing. SIAM Review, 2017, 59(1): 65–98.
[doi: 10.1137/141000671]
[13] Ashouri AH, Killian W, Cavazos J, Palermo G, Silvano C. A survey on compiler autotuning using machine learning. ACM Computing
Surveys (CSUR), 2018, 51(5): 96. [doi: 10.1145/3197978]
[14] Bodin F, Kisuki T, Knijnenburg P, O’Boyle M, Rohou E. Iterative compilation in a non-linear optimisation space. 1998. https://liacs.
leidenuniv.nl/assets/PDF/TechRep/tr98-16.pdf
[15] Bei ZD, Yu ZB, Zhang HL, Xiong W, Xu CZ, Eeckhout L, Feng SZ. RFHOC: A random-forest approach to auto-tuning Hadoop’s
configuration. IEEE Trans. on Parallel and Distributed Systems, 2016, 27(5): 1470–1483. [doi: 10.1109/TPDS.2015.2449299]
[16] Chen TQ, Moreau T, Jiang ZH, Zheng LM, Yan E, CowanM, Shen HC, Wang LY, Hu YW, Ceze L, Guestrin C, Krishnamurthy A. TVM:
An automated end-to-end optimizing compiler for deep learning. In: Proc. of the 13th USENIX Conf. on Operating Systems Design and
Implementation. Carlsbad: USENIX Association, 2018. 579–594.
[17] Mahgoub A, Wood P, Ganesh S, Mitra S, Gerlach W, Harrison T, Meyer F, Grama A, Bagchi S, Chaterji S. Rafiki: A middleware for
parameter tuning of NoSQL datastores for dynamic metagenomics workloads. In: Proc. of the 18th ACM/IFIP/USENIX Middleware
Conf. Las Vegas: ACM, 2017. 28–40. [doi: 10.1145/3135974.3135991]
[18] Ashouri A, Bignoli A, Palermo G, Silvano C, Kulkarni S, Cavazos J. MiCOMP: Mitigating the compiler phase-ordering problem using
optimization sub-sequences and machine learning. ACM Trans. on Architecture and Code Optimization (TACO), 2017, 14(3): 29. [doi:
10.1145/3124452]
[19] Liu HZ, Luo J, Li Y, Wu ZH. Iterative compilation optimization based on metric learning and collaborative filtering. ACM Trans. on
Architecture and Code Optimization (TACO), 2021, 19(1): 2. [doi: 10.1145/3480250]
[20] Theodoridis T, Grosser T, Su ZD. Understanding and exploiting optimal function inlining. In: Proc. of the 27th ACM Int’l Conf. on