Big Data Analytics Big Data Analytics B. Distributed Storage / B.2 Partioning of Relational Databases Lars Schmidt-Thieme Information Systems and Machine Learning Lab (ISMLL) Institute for Computer Science University of Hildesheim, Germany Lars Schmidt-Thieme, Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Germany 1 / 33 Big Data Analytics Syllabus Tue. 9.4. (1) 0. Introduction Tue. 16.4. Tue. 23.4. Tue. 30.4. (2) (3) (4) A. Parallel Computing A.1 Threads A.2 Message Passing Interface (MPI) A.3 Graphical Processing Units (GPUs) Tue. 7.5. Tue. 14.5. Tue. 21.5. (5) (6) (7) B. Distributed Storage B.1 Distributed File Systems B.2 Partioning of Relational Databases B.3 NoSQL Databases Tue. 28.5. Tue. 4.6. Tue. 11.6. Tue. 18.6. (8) (9) (10) C. Distributed Computing Environments C.1 Map-Reduce Pentecoste Break — C.2 Resilient Distributed Datasets (Spark) C.3 Computational Graphs (TensorFlow) Tue. 25.6. Tue. 2.7. (11) (12) D. Distributed Machine Learning Algorithms D.1 Distributed Stochastic Gradient Descent D.2 Distributed Matrix Factorization Tue. 9.7. (13) Questions and Answers Lars Schmidt-Thieme, Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Germany 1 / 33 Big Data Analytics Outline 1. Introduction 2. Horizontal Partitioning 3. Horizontal Partitioning / Parallel Query Processing 4. Vertical Partitioning 5. Sparse Data in Relational Databases Lars Schmidt-Thieme, Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Germany 1 / 33 Big Data Analytics 1. Introduction Outline 1. Introduction 2. Horizontal Partitioning 3. Horizontal Partitioning / Parallel Query Processing 4. Vertical Partitioning 5. Sparse Data in Relational Databases Lars Schmidt-Thieme, Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Germany 1 / 33 Big Data Analytics 1. Introduction Replication and Partitioning I traditionally, relational da