Dataset Popularity Prediction for Caching of CMS Big Data
The Compact Muon Solenoid (CMS) experiment at the European Organization for Nuclear Research (CERN) deploys its data collections, simulation and analysis activities on a distributed computing infrastructure involving more than 70 sites worldwide. The historical usage data recorded by this large infrastructure is a rich source of information for system tuning and capacity planning. In this paper we investigate how to leverage machine learning on this huge amount of data in order to discover patterns and correlations useful to enhance the overall efficiency of the distributed infrastructure in terms of CPU utilization and task completion time. In particular we propose a scalable pipeline of components built on top of the Spark engine for large-scale data processing, whose goal is collecting from different sites the dataset access logs, organizing them into weekly snapshots, and training, on these snapshots, predictive models able to forecast which datasets will become popular over time. The high accuracy achieved indicates the ability of the learned model to correctly separate popular datasets from unpopular ones. Dataset popularity predictions are then exploited within a novel data caching policy, called PPC (Popularity Prediction Caching). We evaluate the performance of PPC against popular caching policy baselines like LRU (Least Recently Used). The experiments conducted on large traces of real dataset accesses show that PPC outperforms LRU reducing the number of cache misses up to 20% in some sites.
KeywordsMachine learning Big data Dataset popularity Classification Caching strategies
Unable to display preview. Download preview PDF.
The first author thanks Tommaso Boccali for his help with the references to Physics subjects and the scientific affiliation with INFN and CERN, as well as Luca Menichetti for his valuable assistance with the Hadoop cluster at CERN and the related software frameworks. The authors thank the CMS experiment for the access to the computing resources and the monitoring logs, and the members of the CMS publications committee.
- 3.Baraglia, R., Castillo, C., Donato, D., Nardini, F.M., Perego, R., Silvestri, F.: Aging effects on query flow graphs for query suggestion. In: Proceedings of the 18th ACM Conference on Information and Knowledge Management, CIKM 2009, Hong Kong, China, November 2–6, 2009, pp 1947–1950 (2009)Google Scholar
- 7.Bonacorsi, D., Kuznetsov, V., Wildish, T., Giommi, L.: Exploring patterns and correlations in CMS computing operations data with big data analytics techniques. In: Proceedings, International Symposium on Grids and Clouds 2015 (ISGC2015): Taipei, Taiwan, March 15–20, 2015, vol. ISGC2015, p 008 (2015)Google Scholar
- 11.Caruana, R., Karampatziakis, N., Yessenalina, A.: An empirical evaluation of supervised learning in high dimensions. In: Machine Learning, Proceedings of the Twenty-Fifth International Conference (ICML 2008), Helsinki, Finland, June 5–9, 2008, pp 96–103 (2008)Google Scholar
- 12.Čerkasova, L., Laboratories, H.P.: Improving WWW Proxies Performance with Greedy-Dual-Size-Frequency Caching Policy. HP Laboratories technical report. Hewlett-Packard Laboratories (1998)Google Scholar
- 13.Chatrchyan, S., et al.: The CMS experiment at the CERN LHC. JINST 3, S08,004 (2008)Google Scholar
- 15.Daruru, S., Marin, N.M., Walker, M., Ghosh, J.: Pervasive parallelism in data mining: dataflow solution to co-clustering large and sparse Netflix data. In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Paris, France, June 28–July 1, 2009, pp 1115–1124 (2009)Google Scholar
- 17.Dietterich, T.G.: Ensemble methods in machine learning. In: Multiple Classifier Systems, First International Workshop, Cagliari, Italy, June 21–23, 2000, Proceedings, pp 1–15 (2000)Google Scholar
- 22.Hushchyn, M., Charpentier, P., Ustyuzhanin, A.: Disk storage management for LHCb based on Data Popularity estimator. CoRR. arXiv:1510.00132 (2015)
- 23.Kavulya, S., Tan, J., Gandhi, R., Narasimhan, P.: An analysis of traces from a production MapReduce cluster. In: 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing, CCGrid 2010, 17–20 May 2010, Melbourne, Victoria, Australia, pp 94–103 (2010)Google Scholar
- 25.Lucchese, C., Nardini, F.M., Orlando, S., Perego, R., Tonellotto, N.: Speeding up document ranking with rank-based features. In: Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval, Santiago, Chile, August 9–13, 2015, pp 895–898 (2015)Google Scholar
- 26.Meng, X., Bradley, J., Yavuz, B., Sparks, E., Venkataraman, S., Liu, D., Freeman, J., Tsai, D., Amde, M., Owen, S., Xin, D., Xin, R., Franklin, M.J., Zadeh, R., Zaharia, M., Talwalkar, A.: MLlib: machine learning in apache spark. J. Mach. Learn. Res. 17(1), 1235–1241 (2016)MathSciNetzbMATHGoogle Scholar
- 28.Meoni, M., Kuznetsov, V., Boccali, T., Menichetti, L.M., Rumševičius, J.: Exploiting apache spark platform for CMS computing analytics. To appear (2017)Google Scholar