Abstract
The increasing complexity and unpredictability of emerging applications makes it challenging for multi-processor system-on-chips to satisfy their performance requirements while keeping power consumption within bounds. In order to tackle this problem, the research community has focused on developing dynamic resource managers that aim to optimize runtime parameters, such as clock frequency, voltage and task mapping. There is a large diversity in the approaches proposed in this context, but a class of resource managers that has gained traction recently is that of reinforcement learning-based controllers. In this paper we propose CoLeCTs, a resource manager that enhances the state-of-the-art resource manager SOSA by employing a joint reward assignment function and enabling collaborative information exchange among multiple learning agents. In this manner we tackle the suboptimal determination of local performance targets for heterogeneous applications and allow cooperative decision making for the learning agents. We evaluate and quantify the benefits of our approach via trace-based simulations.
We acknowledge the financial support from the DFG Grant HE4584/7-2.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Askarizade Haghighi, M., Maeen, M., Haghparast, M.: An energy-efficient dynamic resource management approach based on clustering and meta-heuristic algorithms in cloud computing IaaS platforms. Wirel. Pers. Commun. 104(4), 1367–1391 (2019)
Beckmann, N., Sanchez, D.: Maximizing cache performance under uncertainty. In: 2017 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 109–120. IEEE (2017)
Cobham Gaisler AB: GRLIB IP Library User’s Manual (2022). https://www.gaisler.com/products/grlib/grlib.pdf, version 2022.2
Costero, L., Iranfar, A., Zapater, M., Igual, F.D., Olcoz, K., Atienza, D.: MAMUT: multi-agent reinforcement learning for efficient real-time multi-user video transcoding. In: 2019 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 558–563. IEEE (2019)
Costero, L., Iranfar, A., Zapater, M., Igual, F.D., Olcoz, K., Atienza, D.: Resource management for power-constrained HEVC transcoding using reinforcement learning. IEEE Trans. Parallel Distrib. Syst. 31(12), 2834–2850 (2020)
Cucker, F., Smale, S.: Emergent behavior in flocks. IEEE Trans. Autom. Control 52(5), 852–862 (2007)
Deng, Q., Meisner, D., Bhattacharjee, A., Wenisch, T.F., Bianchini, R.: CoScale: coordinating CPU and memory system DVFS in server systems. In: 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture, pp. 143–154. IEEE (2012)
Donyanavard, B., et al.: SOSA: self-optimizing learning with self-adaptive control for hierarchical system-on-chip management. In: Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture, pp. 685–698 (2019)
Donyanavard, B., Mück, T., Sarma, S., Dutt, N.: SPARTA: runtime task allocation for energy efficient heterogeneous manycores. In: 2016 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ ISSS), pp. 1–10. IEEE (2016)
Dutt, N., Kurdahi, F.J., Ernst, R., Herkersdorf, A.: Conquering MPSoC complexity with principles of a self-aware information processing factory. In: Proceedings of the Eleventh IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis, pp. 1–4 (2016)
Gupta, U., et al.: Adaptive performance prediction for integrated GPUs. In: 2016 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), pp. 1–8. ACM (2016)
Kim, W., Gupta, M.S., Wei, G.Y., Brooks, D.: System level analysis of fast, per-core DVFS using on-chip switching regulators. In: 2008 IEEE 14th International Symposium on High Performance Computer Architecture, pp. 123–134. IEEE (2008)
Kim, Y.G., Wu, C.J.: Autoscale: energy efficiency optimization for stochastic edge inference using reinforcement learning. In: 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO), pp. 1082–1096. IEEE (2020)
LLVM: LLVM “test-suite” repository (2022). https://github.com/llvm/llvm-test-suite/tree/main/SingleSource/Benchmarks/Stanford
Ma, Y., Zhou, J., Chantem, T., Dick, R.P., Wang, S., Hu, X.S.: Online resource management for improving reliability of real-time systems on “Big-Little’’ type MPSoCs. IEEE Trans. Comput.-Aided Des. Integrated Circuits Syst. 39(1), 88–100 (2018)
Maity, B., Donyanavard, B., Dutt, N.: Self-aware memory management for emerging energy-efficient architectures. In: 2020 11th International Green and Sustainable Computing Workshops (IGSC), pp. 1–8. IEEE (2020)
Maurer, F., Donyanavard, B., Rahmani, A.M., Dutt, N., Herkersdorf, A.: Emergent control of MPSoC operation by a hierarchical supervisor/reinforcement learning approach. In: 2020 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 1562–1567. IEEE (2020)
del Mestre Martins, A.L., da Silva, A.H.L., Rahmani, A.M., Dutt, N., Moraes, F.G.: Hierarchical adaptive multi-objective resource management for many-core systems. J. Syst. Architect. 97, 416–427 (2019)
Mishra, N., Imes, C., Lafferty, J.D., Hoffmann, H.: CALOREE: learning control for predictable latency and low energy. ACM SIGPLAN Not. 53(2), 184–198 (2018)
Moazzemi, K., Maity, B., Yi, S., Rahmani, A.M., Dutt, N.: HESSLE-FREE: heterogeneous systems leveraging fuzzy control for runtime resource management. ACM Trans. Embed. Comput. Syst. (TECS) 18(5s), 1–19 (2019)
Mück, T., Donyanavard, B., Moazzemi, K., Rahmani, A.M., Jantsch, A., Dutt, N.: Design methodology for responsive and robust MIMO control of heterogeneous multicores. IEEE Trans. Multi-Scale Comput. Syst. 4(4), 944–951 (2018)
Pal, S., Amarnath, A., Feng, S., O’Boyle, M., Dreslinski, R., Dubach, C.: SparseAdapt: runtime control for sparse linear algebra on a reconfigurable accelerator. In: 54th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO-54, pp. 1005–1021 (2021)
Panait, L., Luke, S.: Cooperative multi-agent learning: the state of the art. Auton. Agent Multi-Agent Syst. 11(3), 387–434 (2005)
Pothukuchi, R.P., Ansari, A., Voulgaris, P., Torrellas, J.: Using multiple input, multiple output formal control to maximize resource efficiency in architectures. In: 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA), pp. 658–670. IEEE (2016)
Pothukuchi, R.P., Pothukuchi, S.Y., Voulgaris, P., Torrellas, J.: Yukta: multilayer resource controllers to maximize efficiency. In: 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA), pp. 505–518. IEEE (2018)
Rahmani, A.M., et al.: SPECTR: formal supervisory control and coordination for many-core systems resource management. In: Proceedings of the Twenty-Third International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 169–183 (2018)
Rahmani, A.M., Haghbayan, M.H., Miele, A., Liljeberg, P., Jantsch, A., Tenhunen, H.: Reliability-aware runtime power management for many-core systems in the dark silicon era. IEEE Trans. Very Large Scale Integration (VLSI) Syst. 25(2), 427–440 (2016)
Ramadge, P.J., Wonham, W.M.: The control of discrete event systems. Proc. IEEE 77(1), 81–98 (1989)
da Rosa, T.R., Larréa, V., Calazans, N., Moraes, F.G.: Power consumption reduction in MPSoCs through DFS. In: 2012 25th Symposium on Integrated Circuits and Systems Design (SBCCI), pp. 1–6. IEEE (2012)
Sadighi, A., et al.: Design methodologies for enabling self-awareness in autonomous systems. In: 2018 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 1532–1537. IEEE (2018)
Tan, M.: Multi-agent reinforcement learning: independent vs. cooperative agents. In: Proceedings of the Tenth International Conference on Machine Learning, pp. 330–337 (1993)
Xun, L., Tran-Thanh, L., Al-Hashimi, B.M., Merrett, G.V.: Optimising resource management for embedded machine learning. In: 2020 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 1556–1561. IEEE (2020)
Zeppenfeld, J., Herkersdorf, A.: Applying autonomic principles for workload management in multi-core systems on chip. In: Proceedings of the 8th ACM International Conference on Autonomic Computing, pp. 3–10 (2011)
Acknowledgement
We thank our IPF project partners at TU Braunschweig and UC Irvine, Rolf Ernst, Fadi Kurdahi, Nikil Dutt and their teams, as well as our colleagues at TUM for their valuable feedback and suggestions during our discussions.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zyla, K., Maurer, F., Wild, T., Herkersdorf, A. (2023). CoLeCTs: Cooperative Learning Classifier Tables for Resource Management in MPSoCs. In: Goumas, G., Tomforde, S., Brehm, J., Wildermann, S., Pionteck, T. (eds) Architecture of Computing Systems. ARCS 2023. Lecture Notes in Computer Science, vol 13949. Springer, Cham. https://doi.org/10.1007/978-3-031-42785-5_15
Download citation
DOI: https://doi.org/10.1007/978-3-031-42785-5_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-42784-8
Online ISBN: 978-3-031-42785-5
eBook Packages: Computer ScienceComputer Science (R0)