Abstract
In embedded systems, tightly coupled memories (TCMs) are usually shared between multiple masters for the purpose of performance scalability, hardware efficiency and software flexibility. On the one hand, memory sharing improves area utilization, but on the other hand, this can lead to a performance degradation due to an increase in access conflicts. To mitigate the associated performance penalty, access interval prediction (AIP) has been proposed. In a similar fashion to branch prediction, AIP exploits program flow regularity to predict the cycle of the next memory access. We show that this structural similarity allows for adaption of state-of-the-art branch predictors, such as the TAgged GEometric history length (TAGE) branch predictor. Our analysis on memory access traces reveals that the obtained TAGE access interval predictor predicts over 97% of memory accesses outperforming all previously presented AIP schemes.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Haas, S., et al.: A heterogenous SDR MPSoC in 28 nm CMOS for low-latency wireless applications. In: DAC. ACM Press, New York (2017). https://doi.org/10.1145/3061639.3062188
Wittig, R., Pauls, F., Matus, E., Fettweis, G.: Access interval prediction for tightly coupled memory systems. In: Pnevmatikatos, D.N., Pelcat, M., Jung, M. (eds.) SAMOS 2019. LNCS, vol. 11733, pp. 229–240. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-27562-4_16
Wittig, R., Hasler, M., Matus, E., Fettweis, G.: Statistical access interval prediction for tightly coupled memory systems. In: COOL CHIPS. IEEE (2019). https://doi.org/10.1109/CoolChips.2019.8721304
Mittal, S.: A survey of techniques for dynamic branch prediction. Concurr. Comput. Pract. Exp. 31(1), 1–36 (2018)
Seznec, A., Michaud, P.: A case for (partially) TAgged GEometric history length branch prediction. J. Instruct. Level Parall. 8 (2006). https://www.jilp.org/vol8
Guthaus, M.R., Ringenberg, J.S., Ernst, D., Austin, T.M., Mudge, T., Brown, R.B.: MiBench: a free, commercially representative embedded benchmark suite. In: 4th Annual Workshop on Workload Characterization. IEEE (2001)
Schoeberl, M., et al.: T-CREST: time-predictable multi-core architecture for embedded systems. J. Syst. Archit. 61(9), 449–471 (2015). https://doi.org/10.1016/j.sysarc.2015.04.002
Paolieri, M., Quiñones, E., Carzola, F.J., Bernat, G., Valero, M.: Hardware support for WCET analysis of hard real-time multicore systems. In: ISCA. ACM Press, New York (2009). https://doi.org/10.1145/1555754.1555764
Ax, J., et al.: CoreVA-MPSoC: a many-core architecture with tightly coupled shared and local data memories. IEEE Trans. Parallel Distrib. Syst. 29(5), 1030–1043 (2018). https://doi.org/10.1109/TPDS.2017.2785799
Gelado, I., Cabezas, J., Stone, J.E., Patel, S., Navarro, N., Hwu, W.-M.W.: An asymmetric distributed shared memory model for heterogeneous parallel systems. In: ASPLOS. ACM Press, New York (2010). https://doi.org/10.1145/1736020.1736059
Wittig, R., Hasler, M., Matus, E., Fettweis, G.: Queue based memory management unit for heterogeneous MPSoCs. In: DATE. IEEE (2019). https://doi.org/10.23919/DATE.2019.8715129
Bates, D., Bradbury, A., Koltes, A., Mullins, R.: Exploiting tightly-coupled cores. In: SAMOS. IEEE (2013). https://doi.org/10.1109/SAMOS.2013.6621138
Rahimi, A., Loi, I., Kakoee, M.R., Benini, L.: A fully-synthesizable single-cycle interconnection network for shared-L1 processor clusters. In: DATE. IEEE (2011). https://doi.org/10.1109/DATE.2011.5763085
Gautschi, M., Rossi, D., Benini, L.: Customizing an open source processor to fit in an ultra-low power cluster with a shared L1 memory. In: GLSVLSI. IEEE (2014). https://doi.org/10.1145/2591513.2591569
Cleary, J., Witten, I.: Data compression using adaptive coding and partial string matching. IEEE Trans. Commun. 32(4), 396–402 (1984)
Michaud, P.: A PPM-like, tag-based branch predictor. J. Instruct. Level Parall. 7, (2005). https://www.jilp.org/vol7
Lowe-Power, J., et al. The gem5 simulator: version 20.0+. (2020). https://arxiv.org/abs/2007.03152
Acknowledgment
This work was funded in part by the German Federal Ministry of Education and Research (BMBF) in the project “E4C” (project number 16ME0426K). We thank the Center for Information Services and High Performance Computing (ZIH) at TU Dresden for generous allocation of compute resources.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Razilov, V., Wittig, R., Matúš, E., Fettweis, G. (2022). Tagged Geometric History Length Access Interval Prediction for Tightly Coupled Memory Systems. In: Orailoglu, A., Reichenbach, M., Jung, M. (eds) Embedded Computer Systems: Architectures, Modeling, and Simulation. SAMOS 2022. Lecture Notes in Computer Science, vol 13511. Springer, Cham. https://doi.org/10.1007/978-3-031-15074-6_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-15074-6_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-15073-9
Online ISBN: 978-3-031-15074-6
eBook Packages: Computer ScienceComputer Science (R0)