Implementasi Deep Learning untuk Prediksi Penggunaan Resource CPU dan RAM pada Server Debian Linux

Firamawati Hia; Ronita Olive Angelie; Lotar Mateus Sinaga

doi:10.31004/riggs.v5i2.11542

Authors

Firamawati Hia Universitas Katolik Santo Thomas Medan
Ronita Olive Angelie Universitas Katolik Santo Thomas Medan
Lotar Mateus Sinaga Universitas Katolik Santo Thomas Medan

DOI:

https://doi.org/10.31004/riggs.v5i2.11542

Keywords:

Deep Learning, Debian Linux, LSTM, Monitoring Server, PyTorch, Prediksi Resource

Abstract

Penelitian ini bertujuan merancang dan mengimplementasikan sistem prediksi penggunaan sumber daya CPU dan RAM pada server Debian Linux menggunakan metode deep learning berbasis Long Short-Term Memory (LSTM). Data monitoring dikumpulkan secara otomatis dan real-time melalui modul collect.py dengan interval lima detik, menghasilkan 1.859 record yang memuat timestamp, persentase CPU, persentase RAM, penggunaan RAM dalam megabyte, dan jumlah proses aktif. Tahap pemrosesan meliputi pengurutan berdasarkan waktu, ekstraksi fitur jam dan hari, normalisasi menggunakan MinMaxScaler, serta pembentukan sequence dengan sliding window sebanyak 30 timestep. Model dibangun menggunakan PyTorch dengan dua lapisan LSTM berukuran 64 dan 32 unit, dilanjutkan lapisan fully connected, ReLU, dan dropout. Pelatihan dilakukan selama 50 epoch menggunakan optimizer Adam dan fungsi loss Mean Squared Error. Hasil pengujian menunjukkan training loss akhir sebesar 0,0140, test loss 0,0310, RMSE 0,1761, dan MAE 0,1426 pada data ternormalisasi. Modul predict.py mampu menghasilkan estimasi CPU dan RAM setiap lima detik serta memberikan klasifikasi kondisi NORMAL atau WASPADA berdasarkan ambang batas yang ditentukan. Sistem dapat berjalan stabil dengan latensi prediksi kurang dari satu detik, sehingga berpotensi mendukung monitoring dan perencanaan kapasitas server secara proaktif. Meskipun demikian, periode pengumpulan data yang singkat dan dominasi kondisi beban rendah membatasi generalisasi model, sehingga pengujian dengan data yang lebih panjang dan bervariasi masih diperlukan.

Downloads

Download data is not yet available.

References

[1] S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.

[2] J. Brownlee, Deep Learning for Time Series Forecasting. Machine Learning Mastery, 2018.

[3] E. Shi et al., "Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms," in Proc. 26th ACM SOSP, 2017, pp. 153–167.

[4] H. Halimatusyadiah, Y. Afrianto, and B. A. Prakosa, "Implementasi LSTM untuk prediksi beban server Raspberry Pi sebagai dasar load balancing," Jurnal Informatika & Komputasi, vol. 8, no. 1, pp. 45–59, 2025.

[5] F. Pedregosa et al., "Scikit-learn: Machine Learning in Python," Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.

[6] A. Paszke et al., "PyTorch: An Imperative Style, High-Performance Deep Learning Library," in Advances in Neural Information Processing Systems (NeurIPS), vol. 32, 2019.

[7] T. F. Arya, R. F. Rachmad, and A. Affandi, "Prediksi resource requirement menggunakan machine learning stacking pada sistem cloud computing," Jurnal Sistem Informasi Indonesia, 2024.

[8] G. van Rossum and F. L. Drake, Python 3 Reference Manual. Scotts Valley, CA: CreateSpace, 2009.

[9] Psutil contributors, "psutil: Cross-platform process and system monitoring library for Python," Python Package Index, 2023. [Online]. Available: https://pypi.org/project/psutil/

[10] W. Matoussi and T. Hamrouni, "A new temporal locality-based workload prediction approach for SaaS services in a cloud environment," Journal of King Saud University - Computer and Information Sciences, vol. 34, pp. 3973–3987, 2022, doi: 10.1016/j.jksuci.2021.04.008.

[11] P. Nawrocki and P. Osypanka, "Cloud Resource Demand Prediction using Machine Learning in the Context of QoS Parameters," Journal of Grid Computing, vol. 19, art. no. 20, 2021, doi: 10.1007/s10723-021-09561-3.

[12] M. M. Al-Sayed, "Workload Time Series Cumulative Prediction Mechanism for Cloud Resources Using Neural Machine Translation Technique," Journal of Grid Computing, vol. 20, art. no. 16, 2022, doi: 10.1007/s10723-022-09607-0.

[13] S. Tuli, S. S. Gill, P. Garraghan, R. Buyya, G. Casale, and N. R. Jennings, "START: Straggler Prediction and Mitigation for Cloud Computing Environments Using Encoder LSTM Networks," IEEE Transactions on Services Computing, 2023.

[14] C. Tang et al., "Forecasting SQL Query Cost at Twitter," in Proceedings of the 2021 IEEE International Conference on Cloud Engineering (IC2E), 2021, pp. 154–160, doi: 10.1109/IC2E52221.2021.00030.

[15] S. Zou, W. Ji, and J. Huang, "BTP: automatic identification and prediction of tasks in data center networks," Journal of Cloud Computing, vol. 11, art. no. 36, 2022, doi: 10.1186/s13677-022-00312-7.

[16] J.-W. Park, M.-W. Kwon, and T. Hong, "Queue congestion prediction for large-scale high performance computing systems using a hidden Markov model," The Journal of Supercomputing, vol. 78, pp. 12202–12223, 2022, doi: 10.1007/s11227-022-04356-z.

[17] T. Zheng, J. Wan, J. Zhang, and C. Jiang, "Deep Reinforcement Learning-Based Workload Scheduling for Edge Computing," Journal of Cloud Computing, vol. 11, art. no. 3, 2022, doi: 10.1186/s13677-021-00276-0.

[18] H. Wu, J. Xu, J. Wang, and M. Long, "Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting," in Advances in Neural Information Processing Systems, vol. 34, 2021.

[19] Y. Liu, H. Wu, J. Wang, and M. Long, "Non-stationary Transformers: Exploring the Stationarity in Time Series Forecasting," in Advances in Neural Information Processing Systems, vol. 35, 2022.

[20] A. Zeng, M. Chen, L. Zhang, and Q. Xu, "Are Transformers Effective for Time Series Forecasting?," Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, no. 9, pp. 11121–11128, 2023, doi: 10.1609/aaai.v37i9.26317.

[21] K. Stankeviciute, A. M. Alaa, and M. van der Schaar, "Conformal Time-Series Forecasting," in Advances in Neural Information Processing Systems, vol. 34, 2021.

[22] F.-K. Sun, C. Lang, and D. Boning, "Adjusting for Autocorrelated Errors in Neural Networks for Time Series," in Advances in Neural Information Processing Systems, vol. 34, 2021.

[23] G. Woo, C. Liu, D. Sahoo, A. Kumar, and S. Hoi, "ETSformer: Exponential Smoothing Transformers for Time-Series Forecasting," arXiv preprint arXiv:2202.01381, 2022.

[24] A. Setayesh, H. Hadian, and R. Prodan, "An Efficient Online Prediction of Host Workloads Using Pruned GRU Neural Nets," arXiv preprint arXiv:2303.16601, 2023.

[25] Debian Project, "Debian 12 (bookworm) Release Notes," 2023. [Online]. Available: https://www.debian.org/releases/bookworm/releasenotes