Abstract
Due to the complexity of the cloud environment, the optimization of cloud cache has become a research hotspot. Most of the current research focuses on cloud cache replacement and prefetching. However, these methods do not take into account that the size of the cache in cloud scenarios is much smaller than the size of the workload, and an effective cache replacement algorithm does not improve the performance of the cache very well. To address these challenges, we design a cache admission policy for cloud block storage using deep reinforcement learning. By analyzing the request characteristics of the workload, three types of request-related features are constructed, which can accurately mine the current request, request history, and current cache state. Experimental results show that, compared with the industry widely deployed cache algorithm, our method decreases write to cloud cache by 115.12%, while improving the hit rate.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Song, Z., Berger, D.S., Li, K.: Learning relaxed belady for content distribution network caching. In: Proceeding of Symposium on Networked Systems Design and Implementation, pp. 529–544 (2020)
Mattson, R.L., Gecsei, J., Slutz, D.R., Traiger, I.L.: Evaluation techniques for storage hierarchies. IBM Syst. J. 9(2), 78–117 (1970)
Harrison, M.A., Ruzzo, W.L., Ullman, J.D.: Protection in operating systems. Commun. ACM 19(8), 461–471 (1976)
Ipek, E., Mutlu, O., MartÃnez, J.F., Caruana, R.: Self-optimizing memory controllers: a reinforcement learning approach. ACM SIGARCH Comput. Archit. News 36(3), 39–50 (2008)
Fulcher, B.D., Jones, N.S.: Highly comparative feature-based time-series classification. IEEE Trans. Knowl. Data Eng. 26(12), 3026–3037 (2014)
The Documentation of TSFRESH. https://tsfresh.readthedocs.io/en/latest/index.html. Accessed 23 Feb 2022
Deng, L., Yu, D.: Deep learning: methods and applications. Found. Trends Signal Process. 7(3), 197–387 (2014)
Koller, R., Rangaswami, R.: I/O deduplication: utilizing content similarity to improve I/O performance. ACM Trans. Storage 6(3), 1–26 (2010)
Wu, S., Jiang, H., Feng, D.: Improving availability of RAID-structured storage systems by workload outsourcing. IEEE Trans. Comput. 60(1), 64–79 (2010)
Bhatia, E., Chacon, G., Pugsley, S., Teran, E., Gratz, P.V., Jiménez, D.A.: Perceptron-based prefetch filtering. In: Annual International Symposium on Computer Architecture, pp. 1–13 (2019)
Masouros, D., Xydis, S., Soudris, D.: Rusty: runtime system predictability leveraging LSTM neural networks. IEEE Comput. Archit. Lett. 18(2), 103–106 (2019)
Einziger, G., Friedman, R., Manes, B.: TinyLFU: a highly efficient cache admission policy. ACM Trans. Storage 13(4), 1–31 (2017)
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 Singapore Pte Ltd.
About this paper
Cite this paper
Long, F. (2022). A Cache Admission Policy for Cloud Block Storage Using Deep Reinforcement Learning. In: S. Shmaliy, Y., Abdelnaby Zekry, A. (eds) 6th International Technical Conference on Advances in Computing, Control and Industrial Engineering (CCIE 2021). CCIE 2021. Lecture Notes in Electrical Engineering, vol 920. Springer, Singapore. https://doi.org/10.1007/978-981-19-3927-3_46
Download citation
DOI: https://doi.org/10.1007/978-981-19-3927-3_46
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-3926-6
Online ISBN: 978-981-19-3927-3
eBook Packages: EngineeringEngineering (R0)