Accelerating time series motif discovery in the Intel Xeon Phi KNL processor

  • Ivan FernandezEmail author
  • Alejandro Villegas
  • Eladio Gutierrez
  • Oscar Plata


Time series analysis is an important research topic of great interest in many fields. Recently, the Matrix Profile method, and particularly one of its implementations—the SCRIMP algorithm—has become a state-of-the-art approach in this field. This is a technique that brings the possibility of obtaining exact motifs from a time series, which can be used to infer events, predict outcomes, detect anomalies and more. However, the memory-bound nature of the SCRIMP algorithm limits the execution performance in some processor architectures. In this paper, we analyze the SCRIMP algorithm from the performance viewpoint in the context of the Intel Xeon Phi Knights Landing architecture (KNL), which integrates high-bandwidth memory (HBM) modules, and we combine several techniques aimed at exploiting the potential of this architecture. On the one hand, we exploit the multi-threading and vector capabilities of the architecture. On the other hand, we explore how to allocate data in order to take advantage of the available hybrid memory architecture that conjugates both the high-bandwidth 3D-stacked HBM and the DDR4 memory modules. The experimental evaluation shows a performance improvement up to \(190\,\times \) with respect to the sequential execution and that the use of the HBM memory improves performance in a factor up to \(5\,\times \) with respect to the DDR4 memory.


Time series Motif discovery Shared-memory parallelism High-bandwidth memory Intel Xeon Phi KNL 



  1. 1.
    Yu ZG, Anh V (2001) Time series model based on global structure of complete genome. Chaos Solitons Fractals 12(10):1827–1834CrossRefzbMATHGoogle Scholar
  2. 2.
    Yoon CE, O’Reilly O, Bergen KJ, Beroza GC (2015) Earthquake detection through computationally efficient similarity search. Sci Adv 1(11):e1501057CrossRefGoogle Scholar
  3. 3.
    Li L, Su X, Zhang Y, Lin Y, Li Z (2015) Trend modeling for traffic time series analysis: an integrated study. IEEE Trans Intell Transp Syst 16(6):3430–3439CrossRefGoogle Scholar
  4. 4.
    Refit: smart homes and energy demand reduction. Accessed 17 Jan 2018
  5. 5.
    Yeh CCM, Zhu Y, Ulanova L, Begum N, Ding Y, Dau HA, Silva DF, Mueen A, Keogh E (2016) Matrix profile I: all pairs similarity joins for time series: a unifying view that includes motifs, discords and shapelets. In: 16th IEEE International Conference on Data Mining (ICDM), pp 1317–1322Google Scholar
  6. 6.
    Torkamani S, Lohweg V (2017) Survey on time series motif discovery. Wiley Interdiscip Rev Data Min Knowl Discov 7(2):e1199CrossRefGoogle Scholar
  7. 7.
    Chandola V, Banerjee A, Kumar V (2009) Anomaly detection: a survey. ACM Comput Surv 41(3):15CrossRefGoogle Scholar
  8. 8.
    Kanarachos S, Christopoulos SRG, Chroneos A, Fitzpatrick ME (2017) Detecting anomalies in time series data via a deep learning algorithm combining wavelets, neural networks and Hilbert transforms. Expert Syst Appl 85:292–304CrossRefGoogle Scholar
  9. 9.
    Lee TJ, Gottschlich J, Tatbul N, Metcalf E, Zdonik S (2018) Greenhouse: a zero-positive machine learning system for time-series anomaly detection. arXiv preprint arXiv:1801.03168
  10. 10.
    Wu J, Zeng W, Yan F (2018) Hierarchical temporal memory method for time-series-based anomaly detection. Neurocomputing 273:535–546CrossRefGoogle Scholar
  11. 11.
    Chiu B, Keogh E, Lonardi S (2003) Probabilistic discovery of time series motifs. In: 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp 493–498Google Scholar
  12. 12.
    McGovern A, Rosendahl DH, Brown RA, Droegemeier KK (2011) Identifying predictive multi-dimensional time series motifs: an application to severe weather prediction. Data Min Knowl Discov 22(1–2):232–258CrossRefGoogle Scholar
  13. 13.
    Yi BK, Faloutsos C (2000) Fast time sequence indexing for arbitrary Lp norms. In: 26th International Conference on Very Large Data Bases (VLDB), pp 385–394Google Scholar
  14. 14.
    Lin J, Keogh E, Wei L, Lonardi S (2007) Experiencing SAX: a novel symbolic representation of time series. Data Min Knowl Discov 15(2):107–144MathSciNetCrossRefGoogle Scholar
  15. 15.
    Miniakhmetov R, Movchan A, Zymbler M (2015) Accelerating time series subsequence matching on the Intel Xeon Phi many-core coprocessor. In: 38th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), pp 1399–1404Google Scholar
  16. 16.
    Chrysos G (2012) Intel Xeon Phi coprocessor (codename Knights Corner). In: IEEE Hot Chips 24 Symposium (HCS), pp 1–31Google Scholar
  17. 17.
    Zhu Y, Yeh CCM, Zimmerman Z, Kamgar K, Keogh E (2018) Matrix profile XI: SCRIMP++: time series motif discovery at interactive speeds. In: 18th IEEE International Conference on Data Mining (ICDM)Google Scholar
  18. 18.
    Zhu Y, Zimmerman Z, Senobari NS, Yeh CCM, Funning G, Mueen A, Brisk P, Keogh E (2016) Matrix profile II: exploiting a novel algorithm and GPUs to break the one hundred million barrier for time series motifs and joins. In: 16th IEEE International Conference on Data Mining (ICDM), pp 739–748Google Scholar
  19. 19.
    Jeffers J, Reinders J, Sodani A (2016) Intel Xeon Phi processor high performance programming: knights, Landing edn. Morgan Kaufmann, BurlingtonGoogle Scholar
  20. 20.
    Loh GH (2008) 3D-stacked memory architectures for multi-core processors. ACM SIGARCH Comput Archit News 36(3):453–464MathSciNetCrossRefGoogle Scholar
  21. 21.
    Rakthanmanon T, Campana B, Mueen A, Batista G, Westover B, Zhu Q, Zakaria J, Keogh E (2012) Searching and mining trillions of time series subsequences under dynamic time warping. In: 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp 262–270Google Scholar
  22. 22.
    Superserver 5038k-i specs. Accessed 17 Jan 2018
  23. 23.
  24. 24.
    Sodani A, Gramunt R, Corbal J, Kim HS, Vinod K, Chinthamani S, Hutsell S, Agarwal R, Liu YC (2016) Knights landing: second-generation Intel Xeon Phi product. IEEE Micro 36(2):34–46CrossRefGoogle Scholar
  25. 25.
    Khaldi D, Chapman B (2016) Towards automatic HBM allocation using LLVM: a case study with knights landing. In: Third Workshop on the LLVM Compiler Infrastructure in HPC (LLVM-HPC). IEEE, pp 12–20Google Scholar
  26. 26.
    Feautrier P (1988) Array expansion. In: 2nd ACM International Conference on Supercomputing (ICS), pp 429–441Google Scholar
  27. 27.
    Gutiérrez E, Plata O, Zapata EL (2000) A compiler method for the parallel execution of irregular reductions in scalable shared memory multiprocessors. In: 14th International Conference on Supercomputing (ICS), pp 78–87Google Scholar
  28. 28.
    Intel VTune website. Accessed 17 Jan 2018
  29. 29.
    Intel Advisor website. Accessed 17 Jan 2018
  30. 30.
    Ghose S, Hsieh K, Boroumand A, Ausavarungnirun R, Mutlu O (2018) Enabling the adoption of processing-in-memory: challenges, mechanisms, future research directions. arXiv preprint arXiv:1802.00320

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Computer ArchitectureUniversity of MálagaMálagaSpain

Personalised recommendations