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

Firamawati Hia Hia; Ronita Olive Angelie; Lotar Mateus Sinaga

doi:10.31004/riggs.v5i2.11453

Authors

Firamawati Hia 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.v5i2.11453

Keywords:

Deep Learning, Debian Linux, LSTM, Monitoring Server, Time Management, PyTorch

Abstract

Penelitian ini mengimplementasikan metode deep learning berbasis Long Short-Term Memory (LSTM) untuk memprediksi penggunaan sumber daya CPU dan RAM pada server Debian Linux secara real-time. Data monitoring dikumpulkan otomatis setiap lima detik melalui modul collect.py dan menghasilkan 1.859 rekaman yang memuat timestamp, persentase CPU, persentase RAM, jumlah RAM terpakai, dan jumlah proses aktif. Data diproses melalui pengurutan waktu, ekstraksi fitur jam dan hari, normalisasi MinMaxScaler, serta pembentukan sequence menggunakan sliding window 30 timestep. Model dibangun dengan PyTorch menggunakan dua lapisan LSTM berukuran 64 dan 32 unit, diikuti lapisan fully connected, aktivasi ReLU, dan dropout. Pelatihan dilakukan selama 50 epoch dengan optimizer Adam dan fungsi kerugian Mean Squared Error. Hasil evaluasi akhir menunjukkan training loss 0,0140, test loss 0,0310, RMSE 0,1761, dan MAE 0,1426 pada skala ternormalisasi. Modul predict.py mampu menghasilkan prediksi CPU dan RAM setiap lima detik serta memberikan status NORMAL atau WASPADA berdasarkan ambang CPU 80% dan RAM 85%. Hasil ini menunjukkan bahwa LSTM dapat diterapkan sebagai dasar monitoring prediktif untuk mendukung tindakan proaktif administrator. Meskipun demikian, data dikumpulkan dalam durasi relatif singkat dan didominasi beban rendah, sehingga pengujian pada periode lebih panjang dan variasi beban yang lebih luas tetap diperlukan sebelum sistem digunakan pada lingkungan produksi. Sistem juga mempertahankan arsitektur modular agar mudah dipelihara dan dikembangkan.

Downloads

Download data is not yet available.

References

[1] M. E. Karim, M. M. S. Maswood, S. Das, and A. G. Alharbi, “BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine,” IEEE Access, vol. 9, pp. 131476–131495, 2021, doi: 10.1109/ACCESS.2021.3113714.

[2] S. Ouhame, Y. Hadi, and A. Ullah, “An Efficient Forecasting Approach for Resource Utilization in Cloud Data Center Using CNN-LSTM Model,” Neural Computing and Applications, vol. 33, no. 16, pp. 10043–10055, 2021, doi: 10.1007/s00521-021-05770-9.

[3] J. Bi, S. Li, H. Yuan, and M. Zhou, “Integrated Deep Learning Method for Workload and Resource Prediction in Cloud Systems,” Neurocomputing, vol. 424, pp. 35–48, 2021, doi: 10.1016/j.neucom.2020.11.011.

[4] J. Kumar, A. K. Singh, and R. Buyya, “Self Directed Learning Based Workload Forecasting Model for Cloud Resource Management,” Information Sciences, vol. 543, pp. 345–366, 2021, doi: 10.1016/j.ins.2020.07.012.

[5] B. Lim, S. Ö. Arik, N. Loeff, and T. Pfister, “Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting,” International Journal of Forecasting, vol. 37, no. 4, pp. 1748–1764, 2021, doi: 10.1016/j.ijforecast.2021.03.012.

[6] 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, pp. 22419–22430, 2021.

[7] H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, “Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 12, pp. 11106–11115, 2021, doi: 10.1609/aaai.v35i12.17325.

[8] M. Xu, C. Song, H. Wu, S. S. Gill, K. Ye, and C. Xu, “esDNN: Deep Neural Network Based Multivariate Workload Prediction Approach in Cloud Computing Environments,” ACM Transactions on Internet Technology, vol. 22, no. 3, 2022, doi: 10.1145/3524114.

[9] T. Zhou, Z. Ma, Q. Wen, X. Wang, L. Sun, and R. Jin, “FEDformer: Frequency Enhanced Decomposed Transformer for Long-Term Series Forecasting,” in Proceedings of the 39th International Conference on Machine Learning, PMLR, vol. 162, pp. 27268–27286, 2022.

[10] 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, vol. 16, no. 1, pp. 615–627, 2023, doi: 10.1109/TSC.2021.3129897.

[11] 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.

[12] Y. Nie, N. H. Nguyen, P. Sinthong, and J. Kalagnanam, “A Time Series Is Worth 64 Words: Long-Term Forecasting with Transformers,” in International Conference on Learning Representations, 2023.

[13] H. Wu, T. Hu, Y. Liu, H. Zhou, J. Wang, and M. Long, “TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis,” in International Conference on Learning Representations, 2023.

[14] M. Daraghmeh, A. Agarwal, and Y. Jararweh, “A Multilevel Learning Model for Predicting CPU Utilization in Cloud Data Centers,” in 2023 IEEE International Conference on Dependable, Autonomic and Secure Computing and Related Conferences, 2023, doi: 10.1109/DASC/PiCom/CBDCom/Cy59711.2023.10361305.

[15] 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.

[16] R. Karthikeyan, V. Balamurugan, R. Cyriac, and B. Sundaravadivazhagan, “COSCO2: AI-Augmented Evolutionary Algorithm Based Workload Prediction Framework for Sustainable Cloud Data Centers,” Transactions on Emerging Telecommunications Technologies, vol. 34, no. 1, e4652, 2023, doi: 10.1002/ett.4652.

[17] S. Simaiya, U. K. Lilhore, Y. K. Sharma, K. B. V. Brahma Rao, V. V. R. Maheswara Rao, A. Baliyan, A. Bijalwan, and R. Alroobaea, “A Hybrid Cloud Load Balancing and Host Utilization Prediction Method Using Deep Learning and Optimization Techniques,” Scientific Reports, vol. 14, art. 1337, 2024, doi: 10.1038/s41598-024-51466-0.